RaspBerry Camera Guide

Download

Page 1

THE OFFICIAL RASPBERRY PI

CAMERA

GUIDE

FOR CAMERA MODULE & HIGH QUALITY CAMERA

Page 2

2 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 3

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 4

First published in 2020 by Raspberry Pi Trading Ltd, Maurice Wilkes Building,

St.John'sInnovation Park, Cowley Road, Cambridge, CB4 0DS

Publishing Director: Russell Barnes • Editor: Phil King

Contributors: Dan Aldred, Wesley Archer, Jody Carter, PJ Evans,

Richard Hayler & family, James Singleton, Rob Zwetsloot,

the Raspberry Pi Foundation Education Team

Design: Critical Media

CEO: Eben Upton

ISBN: 978-1-912047-52-9

The publisher and contributors accept no responsibility in respect of any omissions

or errors relating to goods, products or services referred to or advertised in this book.

Except where otherwise noted, the content of this book is licensed under a Creative

Commons Attribution-NonCommercial-ShareAlike 3.0 Unported

(CC BY-NC-SA 3.0)

4 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 5

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Welcome to

The Official Raspberry

Pi Camera Guide

ne of the most popular add-ons for the Raspberry Pi, the ofcial Camera Module –

or the new High Quality Camera – turns your favourite single-board computer into

succeeded by the higher-spec v2 in April 2016. The High Quality Camera was launched in April 2020, offers Ultra HD image resolution, and enables you to attach any C- or CS-mount lens.

and videos from the command line and writing Python programs to automate the process. We’ll reveal how to create time-lapse and slow-motion videos, before moving on to exciting projects including a Minecraft photo booth, wildlife camera trap, and smart door with video. There are just so many things you can do with a Raspberry Pi camera!

Phil King, Editor

a powerful digital camera. Launched back in 2013, the original Camera Module was

In this book we’ll show you how to get started with your Raspberry Pi camera, taking photos

Page 6

Contents

Chapter 1: Getting started 008

Set up and connect your camera and start taking shots

Chapter 2: Precise camera control 016

Use command-line switches to access camera options and effects

Chapter 3: Time-lapse photography 020

Take photos at regular intervals, then turn the images into a video

Chapter 4: High-speed photography 024

Film dazzling slow-motion clips of exciting events

Chapter 5: Control the camera from Python 028

Use the picamera library to access the camera in Python programs

Chapter 6: Stop-motion and selfies 034

Wire up a physical push-button to take photos

Chapter 7: Flash photography using an LED 040

Add an LED ash to shoot images in low light

Chapter 8: Make a Minecraft photo booth 046

Build a booth in Minecraft that takes photos of the real world

6 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 7

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Chapter 9: Make a spy camera 050

Set up a motion-activated spy camera in your room

Chapter 10: Smart door 054

See who’s at the door and know when the post has arrived

Chapter 11: Car Spy Pi 062

Use ANPR to identify who’s parked on your driveway

Chapter 12: Build a wildlife camera trap 070

Detect and photograph animals in your back garden

Chapter 13: Take your camera underwater 076

Explore the underwater world with your Raspberry Pi and camera

Chapter 14: Install a bird box camera 086

Observe nesting birds without disturbing them

Chapter 15: Live-stream video and stills 092

Stream video and regular stills to a remote computer

Chapter 16: Set up a security camera 102

Protect your home from intruders using motionEyeOS

Chapter 17: Quick reference 108

A guide to the camera hardware, commands, and picamera Python library

Page 8

Chapter 1

Getting started

Find out how to connect your High Quality Camera or

Camera Module, enable it, and take your first shots

n this chapter, we show you how to connect the High Quality Camera or Camera Module

to your Raspberry Pi using the supplied ribbon cable. We will then enable it in Raspbian,

before entering some commands in a Terminal window to start shooting photos and

video. Let’s get started…

01. High Quality Camera

The High Quality Camera (HQ Camera for short) can capture higher-resolution images than

the standard Camera Module. Unlike the latter, it doesn’t have a lens already attached.

Instead, it can be used with any standard C- or CS-mount lens; 6 mm and 16 mm lenses are

available to purchase with the camera to help you get started.

6 mm CS-mount lens

A low-cost 6 mm lens is available for the

HQ Camera. This lens is suitable for basic

photography. It can also be used for macro

photography because it can focus objects at

very short distances.

8 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 9

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

The dust cap should be used when a lens is not

attached, as the camera sensor is sensitive to dust

Supplied with the camera, the C-CS adapter

should be used when attaching a C-mount lens

Use this screw to lock the back focus adjustment

ring in position

This ring can be used to adjust the focus when

using a xed-focus lens, or to alter the focal range

of an attached adjustable-focus lens

This enables you to mount the camera on a

standard tripod: take care not to damage the

ribbon when screwing the tripod into the camera

The camera is supplied with a 20 cm ribbon cable

to connect it to Raspberry Pi’s Camera port, but

longer cables are available if needed

Fitting the lens

The 6 mm lens is a CS-mount device,

so does not need the C-CS adapter ring (see

diagram above). It won’t focus properly if the

adapter is tted – so, if necessary, remove it.

Then rotate the lens clockwise all the way into

the back focus adjustment ring.

Back focus adjustment

ring and lock screw

The back focus adjustment ring should be

screwed in fully for the shortest possible back-

focal length. Use the back focus lock screw to

make sure it does not move out of this position

when adjusting the aperture or focus.

9Chapter 1 Getting started

Page 10

Aperture

To adjust the aperture, hold the

camera with the lens facing away from you.

Turn the middle ring while holding the outer

ring, furthest from the camera, steady. Turn

clockwise to close the aperture and reduce

image brightness. Turn anti-clockwise to open

the aperture. Once you are happy with the light

level, tighten the screw on the side of the lens

to lock the aperture.

Focus

First, lock the inner focus ring, labelled

‘NEAR FAR’, in position by tightening its

screw. Now hold the camera with the lens

facing away from you. Hold the outer two

rings of the lens and turn them both clockwise

until the image comes into focus – it’ll take four or ve whole turns. To adjust focus, turn

the outer two rings clockwise to focus on a

nearby object. Turn them anti-clockwise to

focus on a distant object. You may nd you

need to adjust the aperture again after this.

16 mm C-mount lens

The 16 mm lens provides a higher-quality image than the 6 mm lens. It has a narrow

angle of view which is more suited to

viewing distant objects.

Fitting the

C-CS adapter

Ensure the C-CS adapter that comes with the

HQ Camera is tted to the 16 mm lens. The

lens is a C-mount device, so it has a longer

back focus than the 6 mm lens and therefore requires the adapter.

10 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 11

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Fitting the lens

to the camera

Rotate the 16 mm lens and C-CS adapter

clockwise all the way into the back focus

adjustment ring.

Back focus adjustment

ring and lock screw

The back focus adjustment ring should be

screwed in fully. Use the back focus lock screw to

make sure it does not move out of this position

when adjusting the aperture or focus.

Aperture

To adjust the aperture, hold the camera

with the lens facing away from you. Turn the

inner ring, closest to the camera, while holding

the camera steady. Turn clockwise to close the

aperture and reduce image brightness. Turn anti-

clockwise to open the aperture. When happy with

the light level, tighten the screw on the side of

the lens to lock the aperture into position.

Focus

To adjust focus, hold the camera with the

lens facing away from you. Turn the focus ring,

labelled ‘NEAR FAR’, anti-clockwise to focus

on a nearby object. Turn it clockwise to focus on

a distant object. You may nd you need to adjust

the aperture again after this.

11Chapter 1 Getting started

Page 12

02. Connecting and

using the camera

With your HQ Camera with lens – or your Camera Module – ready, it’s time to connect it to

Raspberry Pi and start to capture some images.

Connect ribbon cable to camera

On the HQ Camera or Camera Module board, you’ll

nd a at plastic connector. Carefully pull the sticking-out

edges until the connector pulls part-way out. Slide the ribbon

cable, with the silver edges downwards and the blue plastic

facing upwards, under the ap you just pulled out, then push the ap gently back into place with a click (Figure 1); it doesn’t

matter which end of the cable you use. If the cable is installed

properly, it will be straight and won’t come out if you give it a

gentle tug; if not, pull the ap out and try again.

Figure 1

Figure 2

12 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 13

Figure 3

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Connect cable to Raspberry Pi

Find the Camera port on Raspberry Pi and pull the plastic ap gently upwards. With

Raspberry Pi positioned so the HDMI port is facing you, slide the ribbon cable in so the silver edges are to your left and the blue plastic to your right (Figure 2), then gently push the ap

back into place. If the cable is installed properly, it’ll be straight and won’t come out if you

give it a gentle tug; if not, pull the ap out and try again.

If using a Raspberry Pi Zero, its Camera port is found on the edge of the board. However,

as it’s a smaller size than the regular one on other Raspberry Pi models, you’ll need a camera

adapter cable to use it.

Enable the camera

Connect the power supply back to Raspberry Pi and let it load Raspbian. Before you

can use the camera, you’ll need to tell Raspberry Pi it has one connected: in the Raspbian

menu, select Preferences, then Raspberry Pi Conguration. When the tool has loaded, click the Interfaces tab, nd the Camera entry in the list, and click on the round radio button to the

left of ‘Enabled’ to switch it on (Figure 3). Click OK, and the tool will prompt you to reboot

your Raspberry Pi. Do so and your camera will be ready to use.

Test the camera

To conrm that your camera is correctly installed, you can use the raspistill tool.

This, along with raspivid for videos, is designed to capture images from the camera using

Raspberry Pi’s command-line interface (CLI). In the Raspbian menu, select Accessories, then

13Chapter 1 Getting started

Page 14

Figure 4

Terminal. A black window with green and blue writing in it will appear (Figure 4): this is the

Terminal, which allows you to access the command-line interface.

To take a test shot, type the following into the Terminal:

raspistill -o test.jpg

As soon as you hit the ENTER key, you’ll see a large picture of what the camera sees

appear on-screen (Figure 5). This is called the live preview and, unless you tell raspistill

otherwise, it will last for ve seconds. After those ve seconds are up, the camera will

capture a single still picture and save it in your home folder under the name test.jpg. If you

want to capture another, type the same command again – but make sure to change the output le name, after the -o, or you’ll save over the top of your rst picture.

More advanced commands

The raspistill command has a list of options so long that it borders on the

intimidating. Don’t worry – you won’t need to learn them all, but there are a few that might be

useful to you, such as:

raspistill -t 15000 -o newpic.jpg

The -t option changes the delay before the picture is taken, from the default ve seconds to whatever time you give it in milliseconds – in this case, you have a full 15 seconds to get

14 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 15

Figure 5

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

your shot arranged perfectly after you press ENTER. You can explore more camera options

in the next chapter, or by referring to Chapter 17.

Rotate the image

If the live preview was upside-down, you’ll need to tell raspistill the camera is

rotated. The Camera Module is designed to have the ribbon cable coming out of the bottom

edge; if it’s coming out of the sides or the top, as with some third-party camera mount accessories, you can rotate the image by 90, 180, or 270 degrees using the -rot switch. For

a camera mounted with the cable coming out of the top, use the following command:

raspistill -rot 180 -o test.jpg

SHOOTING VIDEO

For shooting video, raspivid is what you need. Try it out with this Terminal command:

raspivid -t 10000 -o testvideo.h264

This records a ten-second video (10,000 milliseconds) at the 1920 × 1080 resolution. You can also

shoot slow-mo video at 640 × 480 by using:

raspivid -w 640 -h 480 -fps 90 -t 10000 -o test90fps.h264

You can use VLC to play the videos back – see Chapter 4 for more details.

15Chapter 1 Getting started

Page 16

Chapter 2

Precise camera control

Use command-line switches to access a variety

of camera options and effects

o, you’ve connected the HQ Camera or Camera Module to your Raspberry Pi and

learned how to take still photos and shoot videos from the command line. Now

switches and options available. We’ll also take a look at the raspiyuv command, which sends

its unencoded YUV or RGB output directly from the camera component to a le.

is the preview window that appears by default on the screen. First of all, if it’s upside-down, justadd -rot 180 to your raspistill or raspivid command to rotate it. Also, adding -hf and/or -vf will ip the image horizontally and/or vertically.

width. The -p switch takes four parameters: x co-ordinate, y co-ordinate, width, and height. So,

for example:

let’s explore the raspistill and raspivid commands further, including the many

Preview mode

When taking stills or shooting video, one of the rst things you might want to alter

Using the -p switch, you can set the window’s on-screen position, along with its height and

raspistill -o image.jpg -p 20,100,1280,720

…would place the preview window’s top-left corner at co-ordinate (20,100), with a width of

1280 pixels and height of 720 pixels.

Note that if you only want to see a preview without taking a shot, you can simply omit the

-o image.jpg part. The -t switch sets the duration of the preview: you can set it to 0 to make

it stay on screen until you press CTRL+C.

If you want a full-screen preview, this is easily achieved using the -f switch. The -op switch

can be used to adjust the preview’s opacity, from 0 (invisible) to 255 (solid). If you want to

disable the preview window completely, use the -n switch.

16 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 17

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

The preview can be resized and positioned manually, and can also have its opacity adjusted

Camera control options

Like most dedicated digital cameras, the Raspberry Pi HQ Camera or Camera Module

offers a range of options to adjust aspects such as brightness (-br, from 1 to 100), contrast

(-co, -100 to 100), sharpness (-sh, -100 to 100), saturation (-sa, -100 to 100), ISO (-ISO, 100 to

800), and EV compensation (-ev, -10 to 10).

In addition, there are numerous options for exposure mode for shooting in certain scenarios,

akin to the ‘scenes’ found on most digital cameras. Just use the -ex switch followed by one

of the following terms: auto, night, nightpreview, backlight, spotlight, sports, snow,

beach, verylong (long exposure), fixedfps (for video only), antishake, or fireworks.

Automatic white balance can be adjusted by following the -awb switch with one of: off, auto,

sun, cloud, shade, tungsten, fluorescent, incandescent, flash, or horizon.

You can set the shutter speed in microseconds with the -ss switch; the upper limit depends

on the exposure mode and other settings. The metering mode – used for preview and capture

– can be set with -mm to one of the following: average, spot, backlit, or matrix.

There’s also the option of restricting the region of interest to only part of the sensor, using

-roi with parameters for x and y co-ordinates (from top left), width, and height. For example,

to set a ROI halfway across and down the sensor, with quarter-size width and height, you’d use:

-roi 0.5,0.5,0.25,0.25.

Keypress mode

If you’d like to take a still photo at an exact time, rather than having to wait for the

-t switch delay time to elapse, keypress mode is your friend. Just add the -k switch to your

raspistill command, then press the ENTER key to take the shot: it acts like a shutter button.

To exit the procedure, press X followed by ENTER.

17Chapter 2 Precise camera control

Page 18

A multitude of real-time effects may be added to images, including emboss, as shown here

By adding %04d to the end of your le name in the command, you can save every shot you

have taken before aborting:

raspistill -o keypress%04d.jpg -k

Each shot will have a four-digit sequential number added to its le name, so you’ll get

keypress0000.jpg, keypress0001.jpg, keypress0002.jpg, etc. This is a useful technique for

time-lapses using the -tl switch, too: see Chapter 3 for more details.

Image effects

A whole bunch of effects can be added to the camera in real-time, shown in the

preview window. This is achieved by using the -ifx switch followed by one of the following

terms: none, negative, solarise, posterise, sketch, denoise, emboss, oilpaint, hatch,

gpen (graphite sketch effect), pastel, watercolour, film, blur, saturation (adjust colour

saturation of the image), colorswap, washedout, colorpoint, colorbalance, or cartoon.

If you’d like to take monochrome images, you can use the -cfx (colour effect) switch to

achieve this, using the following setting: -cfx 128:128.

To increase contrast between dark and light areas using DRC (dynamic range compression),

use the -drc switch to turn it on/off (it’s off by default).

18 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 19

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Still options

Let’s take a look at some options that are specic to the raspistill command. As

already mentioned, we use -o followed by a le name to output to a le, and the -t switch sets

the shutter delay in milliseconds. For example, to save a photo taken after two seconds, use:

raspistill -t 2000 -o image.jpg

You can set the width and height of the image with -w and -h, each followed by a value – up to

4056 and 3040 (HQ Camera), 3280 and 2464 (Camera Module v2), or 2592 and 1944 (CM v1).

You can also set the quality of the JPEG image, using -q, from 0 to 100 – the latter is almost

completely uncompressed. Alternatively, to save it as a lossless PNG (slower than using JPG),

use -e (encoding) followed by png:

raspistill -o image.png –e png

For a full list of options, see Chapter 17. The raspiyuv command works in a similar fashion

and offers most of the same options, apart from adding EXIF tags, but sends its YUV or RGB output directly from the camera component to le. To use RGB, add the -rgb switch.

Shooting video

The raspivid command is used to shoot video. In this case, the -t switch sets the

duration in milliseconds. The bitrate is set using -b, in bits per second (so, 25Mpbs is -b

25000000), while -fps sets the frame rate – see Chapter 17 for the maximum bitrate and frame rate for your camera model. For example, to shoot ve seconds of video at 1080p

(1920 × 1080), with a bitrate of 15Mbps and frame rate of 30fps, use:

raspivid -t 5000 -b 15000000 -fps 30 -o video.h264

See Chapter 4 for information

on how to shoot slow-

motion footage. Many

OTHER VIDEO FORMATS

transfer the les to a more powerful computer

instead. Whichever method you decide to

use, you will need to install the tools on the

rendering machine; for Raspberry Pi, enter:

sudo apt-get install ffmpeg

This installs the FFmpeg tool which we'll use to convert our images into a video.

To copy the images to a remote machine, you can download them from the web server using

wget or curl. For example:

wget -r -A jpg http://192.168.1.45

Or if you don’t have wget…

curl http://192.168.1.45/frame [0001-0766].jpg -O

WebM is an open video format that can

be displayed directly in most browsers.

However, other video formats are available.

Change the IP address and numbers accordingly.

Make the video

The nal step is to assemble the video. Run the following command to start the

rendering process:

sudo ffmpeg -i /var/www/html/frame%04d.jpg -crf 4 -b:v 10M

/var/www/html/video.webm

When the rendering process has nished, you’ll be able to view the video in your browser.

The default frame rate is 25fps. This compresses three hours of images taken at ten-second

intervals to about 40 seconds of video. You can adjust this with the -framerate command-

line option. The bitrate (-b) has been set high, and the Constant Rate Factor (-crf) has been

kept low, to produce a good-quality video.

Running the

rendering process on a Raspberry Pi. This will take some time, so you may prefer to use a faster machine

23Chapter 3 Time-lapse photography

Page 24

Chapter 4

High-speed photography

All you need to make dazzling slow-motion clips of exciting

events is a Raspberry Pi and HQ Camera or Camera Module

t rst glance it seems counter-intuitive, but in order to create a smooth slow-

motion movie, you need a high-speed camera. Essentially, a movie is just a

matches the original action. A slow-motion clip is produced by recording more frames than



recorded at 24 frames per second (fps), with video frame rates varying between 25 and 29fps

depending on which format/region is involved. So if you record at 50fps and play back at

25fps, the action will appear to be taking place at half the original speed. It’s actually a little

more complicated than that with the use of interlaced frames, but you don’t really need to

consider them here.

collection of still photos, or frames, all played one after the other at a speed that

Clips can now be recorded at a high frame rate

The original software for the Camera Module was limited in terms of the frame rates it could

cope with, but a subsequent update added new functionality so that clips can now be recorded

at up to 90fps (or 120fps on the HQ Camera). There is one slight limitation: high frame rates



So, depending on yor exact hardware setup, a high-speed mode of 90fps may most

consistently be achieved at a lower resolution such as 640×480. This is still good enough to

capture decent-quality images, though.

A quick way of getting started is to pick some everyday objects and record them in

motion. How about a dropped egg hitting a table top? A pull-back toy car crashing through

some Lego blocks? Or even the old favourite of a water balloon bursting? It’s best to do the

last one outside!

24 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 25

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Pick some everyday objects and record them in motion

Once you’ve chosen your subject, you’ll need a way of holding and angling the camera, and

some way of lighting the scene. Neither needs to be sophisticated: a normal desk lamp works



keeping the camera stable at tricky angles. You might also want to invest in a longer cable for



a set of special adapters allows you to extend using a standard HDMI cable.

Avoid unwanted reflections, and fine-tune your video specifications

Note that if using a Camera Module v1, its red LED will illuminate when recording is taking

 disable_camera_led=1 to your /boot/cong.txt

The command for capturing video with the Raspberry Pi camera is raspivid, best run from

a Terminal window. There are a number of command options that you need to specify:

• -fps sets the frame rate.

• -w and -h specify the width and height of the frames. For the fastest frame rates, set this to

640 and 480 respectively.

• -t allows you to set how long to record for. If you’re working by yourself, the easiest way to



plenty of time to start things off manually.

• -o

• -n disables preview mode.

 test.h264:

raspivid -n -w 640 -h 480 -fps 90 -t 5000 -o test.h264

Right: now that you’ve recorded your movie clip, how can you play it back? One easy way is to



recommended software’ image. If it’s not, you can install it with:

sudo apt-get install vlc

The version on Raspberry Pi has some handy features which can be accessed by checking the



by Frame’ button. You can also alter the playback speed to slow things down even further.

25Chapter 4 High-speed photography

Page 26

To extend the project, how about connecting a break-beam IR sensor pair via the GPIO

pins and using these to trigger the camera recording? The Python picamera library (see

Chapter 5) provides full access to the camera’s functions and could be used with your code.

Capturing the clip

Lights

Get your scene lined up and lit, then test how it looks by using the camera preview



raspistill -w 640 -h 480 -t 5000

26 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 27

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Camera

ENTER yet):

raspivid -w 640 -h 480 -fps 90 -t 7000 -o myvid1.h264

Once triggered, this will capture a seven-second clip.

Action

When everything is ready, hit ENTER and then release the car / drop the egg / burst

the balloon. You’ll have footage before and after the event, which can be trimmed with some

post-production editing.

27Chapter 4 High-speed photography

Page 28

Chapter 5

Control the camera from Python

Use the picamera library to access the camera in Python programs

o far, we’ve looked at using the camera from the command line, but what if you want

to control it from a Python program? The picamera library enables you to access all

fully supported by picamera and may freeze when attempting to take multiple shots.) Let’s

take a look at how to use it to shoot stills and videos, alter settings, and add effects.

it’s not present already, you can install it manually. In a Terminal window, enter:

the camera’s features in Python. (Note: at the time of writing, the HQ Camera isn’t yet

Getting started

The picamera library comes pre-installed in the most recent versions of Raspbian. If

sudo apt-get update sudo apt-get install python-picamera python3-picamera

The camera preview can be resized and positioned to your liking

28 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 29

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

With your camera already connected and enabled in Raspberry Pi Conguration, open Programming > Thonny from the Raspbian desktop menu. Create a new le by clicking File > New le. Save it with File > Save, naming it ch5listing1.py. Note: Never name a le picamera.py, as this is the le name for the picamera library itself!

Now enter the code from ch5listing1.py. Save it with CTRL+S and run with F5. The full-

screen camera preview should be shown for ten seconds, and then close. Note: To be able

to see the preview when using VNC for remote access from another computer, open the VNC

Server menu and go to Options > Troubleshooting, then select ‘Enable direct capture mode’.

If the preview appears upside-down, add the line camera.rotation = 180 just above

camera.start_preview(). Other possible rotation values are 90 and 270.

You can alter the transparency level of the preview by entering an alpha value – from 0 to

255 – within the latter command’s brackets; e.g. camera.start_preview(alpha=200).

It’s also possible to change the position and size of the preview. For example, to place its

top corner 50 pixels right and 150 down, and resize it to 1024 × 576:

camera.start_preview(fullscreen=False, window = (50,150,1024,576))

Take a photo

Now let’s take a still photo. We can do this by adding the line:

camera.capture('/home/pi/Desktop/image.jpg')

…just after the sleep in our code, so it looks like ch5listing2.py. Run the code and after a

preview of ve seconds (as set by sleep), it’ll capture a photo as image.jpg. You may the

preview adjust to a different resolution momentarily as the picture is taken. In this example,

the resulting image le will appear on the desktop; double-click its icon to open it.

You can alter the le name and directory path in the code, along with the sleep time.

Remember, though, that it should be at least ve seconds, to give the camera sensor enough

time to adjust its light levels.

Make a loop

The great thing about using Python with the picamera library is that it makes it easy

to use a loop to take a sequence of photos. In Thonny, create a new le and enter the code fromch5listing3.py.

After initiating the camera preview, we add a for loop with a range of 5, so it will run ve times

to take ve photos. The sleep command sets the time between shots, captured using the line:

camera.capture('/home/pi/Desktop/image%s.jpg' % i)

Here, the %s token is replaced by whatever we add after the % following the le name – in this case, the variable i set by our for loop. Note that i will range from 0 to 4, so the images will

29Chapter 5 Control the camera from Python

Page 30

Using the ch5listing5.py code, you can view a loop of all the effects on offer

be saved as image0.jpg, image1.jpg, and so on. Once they’re all taken, the preview will close. In this example, you’ll see the ve les on your desktop; double-click to open them.

You can also use a for loop to alter camera setting levels such as brightness over time. For

more details, see Step 04.

Control camera settings

Brightness is just one of numerous settings available for the camera. Here’s a list of

the main options, along with their default values (and ranges where applicable):

camera.brightness = 50 (0 to 100) camera.sharpness = 0 (-100 to 100) camera.contrast = 0 (-100 to 100) camera.saturation = 0 (-100 to 100) camera.iso = 0 (automatic) (100 to 800) camera.exposure_compensation = 0 (-25 to 25) camera.exposure_mode = 'auto' camera.meter_mode = 'average' camera.awb_mode = 'auto' camera.rotation = 0 camera.hflip = False camera.vflip = False camera.crop = (0.0, 0.0, 1.0, 1.0)

The resolution of the capture is also congurable. For example:

30 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 31

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

camera.resolution = (1024, 768)

The maximum resolution for photos is4056 × 3040 (HQ Camera), 3280 × 2464 (Camera Module v2), or 2592 × 1944 (Camera Module v1). Note: you may need to increase gpu_mem in /boot/cong.txt to achieve full resolution with the Camera Module v2.

Add image effects

Just as when you are using the command line, a wide range of effects can be added to

the camera in real-time, shown in the preview window. The camera.image_effect command

is used to apply a particular image effect. The options are: none (the default), negative,

solarize, sketch, denoise, emboss, oilpaint, hatch, gpen (graphite sketch effect), pastel,

watercolor, film, blur, saturation, colorswap, washedout, posterise, colorpoint,

colorbalance, cartoon, deinterlace1, and deinterlace2.

For instance, to take an image with a colour swap effect, enter the code from ch5listing4.py

and run it.

To loop through thevarious image effects in a preview, run the code from ch5listing5.py. Note

that this uses the camera.annotate_text command to add a text message to the preview; this

can also be applied to captured images (when using the sensor’s full eld of view).

For more details on these effects and other settings, see Chapter 17 or the ofcial picamera

documentation at picamera.readthedocs.io.

Shoot a video

To shoot video footage, we replace the camera.capture() command with

camera.start_recording(), and use camera.stop_recording() to stop. Enter the

example code from ch5listing6.py.

When you run the code, it records ten seconds of video before closing the preview. To play

the resulting le, open a Terminal window from the desktop and enter:

oxplayer video.h264

(Or you can use VLC instead.) Note that it may well play faster than the original frame rate. It’s possible to convert videos to MP4 format and adjust the frame rate using the MP4Box utility

(installed with sudo apt-get install gpac), like so:

MP4Box -add video.h264:fps=30 video.mp4

All of the image effects and most of the camera settings can be applied while shooting video.

You can also turn on video stabilisation, which compensates for camera motion, by adding the

following line to your Python program:

camera.video_stabilization = True

31Chapter 5 Control the camera from Python

Page 32

ch5listing1.py / Python 3

DOWNLOAD

from picamera import PiCamera from time import sleep

camera = PiCamera()

camera.start_preview()

sleep(10)

camera.stop_preview()

ch5listing2.py / Python 3

from picamera import PiCamera from time import sleep

camera = PiCamera()

camera.start_preview()

sleep(5)

camera.capture('/home/pi/Desktop/image.jpg') camera.stop_preview()

ch5listing3.py / Python 3

magpi.cc/cameragit5

from picamera import PiCamera from time import sleep

camera = PiCamera()

camera.start_preview()

for i in range(5):

sleep(5) camera.capture('/home/pi/Desktop/image%s.jpg' % i) camera.stop_preview()

32 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 33

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

ch5listing4.py / Python 3

from picamera import PiCamera from time import sleep

camera = PiCamera()

camera.start_preview() camera.image_effect = 'colorswap'

sleep(5)

camera.capture('/home/pi/Desktop/colorswap.jpg') camera.stop_preview()

ch5listing5.py / Python 3

from picamera import PiCamera from time import sleep

camera = PiCamera()

camera.start_preview()

for effect in camera.IMAGE_EFFECTS:

camera.image_effect = effect camera.annotate_text = "Effect: %s" % effect sleep(5) camera.stop_preview()

ch5listing6.py / Python 3

from picamera import PiCamera from time import sleep

camera = PiCamera()

camera.start_preview() camera.start_recording('/home/pi/video.h264')

sleep(10)

camera.stop_recording() camera.stop_preview()

33Chapter 5 Control the camera from Python

Page 34

Chapter 6

Stop-motion and selfies

Wire up a physical push-button to take photos

ave you been reading the last few chapters and thinking you’d like to take a picture

with a Raspberry Pi camera with less hassle? In this tutorial we’ll show you how to

many projects (for example, time-lapse photography), but in this chapter we are focusing on

stop-motion animation. We also show how to create your own sele stick!

button to Raspberry Pi via a jumper wire, as shown in Figure 1. One side of the button will be

connected to ground (GND); the other is connected to the GPIO 14 pin (but you can choose

your favourite pin). We used a breadboard for our stop-motion animation project, but you could

wire the button directly to the pins (as you’ll be doing for the sele stick later).

take a photo with a click of a button, just like a real camera. This could be useful for

Wire up the button

If you haven’t already switched your Raspberry Pi off, do so now. Next, connect the

Install picamera

That’s all the hardware done. Now it’s time for the software. If you haven’t done so

already in Chapter 5, you’ll need to install the picamera library. In a Terminal window, enter:

sudo apt-get update sudo apt-get install python-picamera python3-picamera

YOU’LL NEED

• Camera Module / HQ Camera

• Push-button

• Breadboard (optional)

• Jumper wires

34 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

• Raspberry Pi case with a hole for the camera

cable (sele stick)

• Long wires (sele stick)

• A stick, slimmetal pole etc. (sele stick)

Page 35

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

You can add the push-button to a breadboard

or wire it directly to the GPIO pins

Figure 1 Connect the button

If for some reason you don’t have GPIO Zero already installed (it has come pre-installed in

Raspbian for some time), do so with:

sudo apt-get install python-gpiozero python3-gpiozero

Stop-motion software

Because we’re focusing on stop-motion for our rst project, we’re using the camera’s

preview mode so that we can set up our shot before we take it, to ensure everything is in the

frame. Then, only when the button is pressed do we save an image le. Each image le will

have a different name based on the date and time at which it is taken. This makes it easy to

assemble all the images from the shoot for post-processing.

The wonderful GPIO Zero library is used to capture the button activity; we simply dene a

function that is run whenever the button is pressed. This function uses the picamera Python

library which allows us to control the camera through code, making all the normal command-

line operations available.

Download or type up the code from ch6listing1.py and either run it through Thonny or the

command line. To quit the program, press CTRL+C.

One button leg is wired to a GPIO pin

(we used GPIO14); the other to GND

35Chapter 6 Stop-motion and seles

Page 36

Other variations

You should be able to use this code as a template to create a program for whatever

photography project you have in mind. For example, you could alter the code so that the

camera takes continuous photos while the button is held down. Or you could add extra buttons

to make a variety of photography modes available.

With this sort of build, you can also start thinking about building a complete, portable,

wirelessly connected Raspberry Pi camera. For this, you can use a case into which you can t

a portable mobile phone battery charger, along with a screen to attach to Raspberry Pi. With

a bit of modication of the code, you can have it always show the preview of the camera on the screen. Want to record video? More modication of the code will allow for video capturing. The only issue you might have with both of these projects is the lack of a ash or built-in light

source, so a well-lit subject would be essential.

Selfie stick

Next, we’ll look at making a sele stick. A lot of people roll their eyes and complain about vanity when it comes to the art of the sele, but we all know it’s nothing like that. New outt? New glasses? Eyeliner wings perfectly symmetrical today? Why not chronicle it? It’s a great condence boost.

You can use a breadboard for a small button, or connect your jumper

wires directly to the pins on a bigger one

36 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 37

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

ch6listing1.py / Python 3



from datetime import datetime from gpiozero import Button import picamera import time

b=Button(14) pc=picamera.PiCamera() running = True

#pc.resolution = (1024, 768) #use this to set the resolution if you dislike the default values

timestamp=datetime.now()

def picture():

pc.capture('pic'+str(timestamp)+'.jpg') #taking the picture

pc.start_preview() #running the preview b.when_pressed=picture

try:

while running: print('Active')#displaying 'active' to the shell time.sleep(1)

#we detect Ctrl+C then quit the program

except KeyboardInterrupt:

pc.stop_preview() running = False

DOWNLOAD

magpi.cc/cameragit6

Our test sele stick is very DIY, but you can use anything as long

as you can attach a Raspberry Pi and have a long enough wire

37Chapter 6 Stop-motion and seles

Page 38

Create your stop-motion scene and use the button to trigger the camera to

take pictures and save them to timestamped le

Our Raspberry Pi-powered sele stick will use a similar hardware and software setup to the

stop-motion animation project. As before, we’re wiring up a push-button to GPIO 14 and GND

pins on Raspberry Pi, but this time we need to attach the jumpers to longer wires to put the

button at the end of the ‘stick’ – we used a spatula, but anything long will do.

Your Raspberry Pi needs to be near to the camera (unless you’ve got an extra-long ribbon

cable). Attach Raspberry Pi in a case to one end of the stick with whatever means you see t

(glue, adhesive putty, string, etc.) and then attach the button.

Add the code

Since the principle is the same – pressing a button to take a photo – we can use the

same code, ch6listing1.py, as for the stop-motion project. This time we don’t need the camera

preview, so you can comment out the line pc.start_preview() if you like, by adding a # to

the start of it.

Try running the code. Pressing the button will take a photo, but you’ll need to practise your

aim so you can get yourself in the frame. As before, we add a timestamp to each picture,

which helps to organise your pictures later and also results in a slight pause in the code, which

at least means you won’t take too many pictures with a slip of the button.

38 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 39

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

39Chapter 6 Stop-motion and seles

Page 40

Chapter 7

Flash photography using an LED

Add an LED flash to shoot images in low light

he Raspberry Pi Camera Module

or HQ Camera works really well in

there’s less light available? Here we show you

how to set up a simple LED ash, which will be triggered each time you take a photo, using the picamera Python library. We also take a look at how to shoot better images in low light when you are not using a ash.

then be triggered each time we capture a still using picamera with the ash mode set to on. To do this, we need to edit the VideoCore GPU default device tree source. First, install device tree compiler with:

Then grab a copy of the default device tree source with:

good lighting conditions, but what if

Download device tree source

Before we can wire up a ash, we need to congure a GPIO pin to use for it. This will

sudo apt-get install device-tree-compiler

wget https://raw.githubusercontent.com/raspberrypi/firmware/

master/extra/dt-blob.dts

Edit the device tree source

Edit the le using your favourite text editor, such as nano:

YOU’LL NEED

• Camera Module / HQ Camera

• White LED

• Resistor

sudo nano dt-blob.dts

40 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 41

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

You’ll need to nd the correct part of the code for the Raspberry Pi model you’re using; for instance, the part for Raspberry Pi 4 is found under pins_4b {.

Here you’ll nd pin_config and pin_defines sections. In the pin_config section, add a

line to congure the GPIO pin (we’re using GPIO 17) that you want to use for the ash:

pin@p17 { function = "output"; termination = "pull_down"; };

Enable flash

Next, we need to associate the pin we added with the ash enable function by editing it in the pin_define section. We simply change absent to internal and add a line with the pin number, so it looks like the following:

pin-define@FLASH_0_ENABLE { type = "internal"; number = <17>; };

Note that it’s the FLASH_0 section that you need to alter: FLASH_1 is for an optional privacy LED to come on after taking a picture, but we won’t bother with that.

The resistor limits the current

owing through the LED

The longest leg of the LED is the

anode: connect it to GPIO 17

Figure 1 Connect a white LED

41Chapter 7 Flash photography using an LED

Page 42

Wire the white LED to GPIO 17 and GND via a low-ohmage resistor

Compile the blob

With the device tree source updated, we now need to compile it into a binary blob,

using the following command in a Terminal window:

dtc -q -I dts -O dtb dt-blob.dts -o dt-blob.bin

This should output nothing. Next, you need to place the new binary on the rst partition of the microSD card. In the case of non-NOOBS Raspbian installs, this is generally /boot, so use:

sudo cp dt-blob.bin /boot/

In you installed Raspbian via NOOBS, however, you’ll need to do the following instead:

sudo mkdir /mnt/recovery sudo mount /dev/mmcblk0p1 /mnt/recovery sudo cp dt-blob.bin /mnt/recovery sudo umount /mnt/recovery sudo rmdir /mnt/recovery

To activate the new device tree conguration, reboot your Raspberry Pi.

42 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 43

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

You need to edit the device tree source to enable a GPIO pin for the ash

Wire up the LED

Connect a white LED – we used a 5 mm one – to your Raspberry Pi as in Figure 1 (previous page). The LED’s anode (long leg) is connected to our ash-enabled pin, GPIO 17. To be sure of the LED not burning out from excess current, you should add a low-ohmage resistor (such as 100 Ω) between the LED’s cathode (short leg) and Raspberry Pi’s GND pin. Depending on the maximum forward voltage of your LED (ours was 3.5V), you may be able to get away without using one, but it’s best to be safe.

If you want to use higher-powered or multiple LEDs, you’ll have to think about powering them via a suitable driver circuit, with a transistor wired to the ash pin. You may also need a separate power supply. Note that, due to the Raspberry Pi camera’s rolling shutter, only an LED or equivalent ash is suitable: you can’t use a xenon ash. Alternative ash/lighting methods include NeoPixel sticks and the LISIPAROI light ring.

Test it out

With the LED connected, we can now test out our ash with a short Python program. In Thonny, create a new le and enter the code from ch7listing1.py. The camera.flash_mode = 'on' line sets the ash to trigger when we issue the capture command below; the LED will light up briey before the image capture, to enable the camera to set the correct exposure level for the extra illumination, before the ash proper is triggered.

If you want the ash to trigger automatically only when it’s dark enough, you can change the

penultimate line of the code to camera.flash_mode = 'auto'.

43Chapter 7 Flash photography using an LED

Page 44

Using a long exposure, you can shoot stills in very dark settings

Low-light photography

In low-light scenarios where you don’t want to use a ash, you can improve capture of images using a few tricks. By setting a high gain combined with a long exposure time, the camera is able to gather the maximum amount of light. Note that since the shutter_speed attribute is constrained by the camera’s frame rate, we need to set a very slow frame rate. The code in ch7listing2.py captures an image with a six-second exposure time: this is the maximum time for the Camera Module v1 – if you have a v2 Camera Module, it can be extended to ten seconds, or much longer for an HQ Camera. The frame rate is set to a sixth of a second, while we set the ISO to 800 for greater exposure. A pause of 30 seconds gives the camera enough time to set gains and measure AWB (auto white balance).

Try running the script in a very dark setting: it may take some time to run, including the 30-second pause and about 20 seconds for the capture itself. Note: if you’re getting a timeout error, you may needto do a full Raspbian upgrade with sudo apt-get update and sudo apt-get dist-upgrade.

The particular camera settings in this script are only useful for very low light conditions: in a less dark environment, the image produced will be heavily overexposed, so you may need to increase the frame rate and lower the shutter speed accordingly.

If the image has a green cast, you’ll need to alter the white balance manually. Turn AWB off with camera.awb_mode = 'off'. Then set the red/blue gains manually;.e.g. camera.awb_gains = (1.5, 1.5).

Even a single LED can provide illumination

for close-up photography

44 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 45

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

ch7listing1.py / Python 3

import picamera

with picamera.PiCamera() as camera:

camera.flash_mode = 'on' camera.capture('foo.jpg')

magpi.cc/cameragit7

ch7listing2.py / Python 3

from picamera import PiCamera from time import sleep from fractions import Fraction

# Set a framerate of 1/6fps, then set shutter # speed to 6s and ISO to 800

camera = PiCamera(resolution=(1280, 720),

framerate=Fraction(1, 6))

camera.shutter_speed = 6000000 camera.iso = 800

# Give the camera a good long time to set gains and # measure AWB (you may wish to use fixed AWB instead)

sleep(30)

camera.exposure_mode = 'off'

# Finally, capture an image with a 6s exposure. Due # to mode switching on the still port, this will take # longer than six seconds

camera.capture('dark.jpg')

DOWNLOAD

45Chapter 7 Flash photography using an LED

Page 46

Chapter 8

Make a Minecraft photo booth

Create a photo booth in Minecraft that takes photos

of the real world. What will you see on your travels?

ot only is Minecraft Pi great fun to play around with, you can also use Python

programming to manipulate the Minecraft world and create various structures

chapter, we’ll be getting Minecraft to trigger the Camera Module or HQ Camera with code when

the player enters a virtual photo booth.

interface). This enables you to connect to Minecraft and program it with Python. You also need to import picamera’s PiCamera class to control the camera, and the time module to add a

small delay between taking each photo.

the Recommended Software tool), then enter an existing world or create a new one. Move the Minecraft window to one side of the screen. You’ll need to use the TAB key to take your

mouse’s focus away from the Minecraft window to move it. This will be needed later when you

switch between the Minecraft and Python windows.

to write the photo booth program.

program with F5. You should see the message ‘Find the photobooth’ appear in the Minecraft

world. This is the rst part of the code. Stop the program running using CTRL+C.

within it. Going beyond this, you can even have it interact with the real world. In this

The rst thing you need to do is import the Minecraft API (application programming

Open Minecraft from the applications menu (if it's not present under Games, install it via

Open Thonny from the applications menu. This will open up the code editor which you’ll use

Enter the code from ch8listing1.py, or download it. Save with CTRL+S and run the

Camera tests

Next, we’ll make sure the camera is set up. We’ve set the camera to show a two-second

preview, so that you can strike your pose and smile before the picture is taken. The image is

stored as a le called sele.jpg in your home directory (/home/pi).

46 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 47

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Now, you need to create a photo booth in the Minecraft environment. This is done manually,

and the booth can be built wherever you want to locate it. Using any block type, build your

photo booth. It can be any shape you like, but it should have at least one block width of free

space inside so that the player can enter.

Once you have created your photo booth, you need to be able to move your player inside and

onto the trigger block. This is the block that the player stands on to run the function that you

wrote in the rst step, which will then trigger the camera. In the Minecraft environment, your

position is given in reference to the x, y, and z axes. Look at the top-right of the window and

you’ll see the x, y, and z co-ordinates of your player – for example, 10.5, 9.0, -44.3. Assuming

you are still in the photo booth, then these are also the x, y, and z co-ordinates of the trigger

block in your booth.

Walk into your photo booth

Note down all three co-ordinates of your camera trigger block. When you’re playing Minecraft,

your program will need to verify that you areinside the photo booth. If you are, then it will

trigger the take_the_pic function and take a picture with the camera. To do this, Minecraft

needs to know where you are in the world.

47Chapter 8 Make a Minecraft photo booth

Page 48

Construct your photo booth however

you wish; just make sure the code

knows where it lives

Steve is your ‘shutter’ in the

Minecraft world: move him

to the booth to take a photo

Place the booth anywhere in your world.

Give it a special room in your house, or use

it as a trap to see if someone is in your world

To nd your position, you use the code x, y, x = mc.player.getPos(). This saves the x, y, and z position of your player into the variables x, y, and z. You can then use print(x) to

print the x value, or print(x, y, z) to see them all if you wish, by adding it to the code. Now

you know the position of the player, you can test to see if they’re in the photo booth.

At this point we have a photo booth, the co-ordinates of the trigger block, and code to

control the camera and take a picture. The next part of the code is to test whether the program

knows when you’re in the photo booth. To do this, we create a loop which checks if your

player’s co-ordinates match the trigger block co-ordinates. If they do, then you’re standing in the photo booth. For this, we use a simple if statement, which is known as a conditional.

Change the if line in the code to ensure the co-ordinates you enter are those of your photo

booth. Save and run your code to test it: walk into your photo booth and you should see the

message ‘You are in the photobooth!’ in the Minecraft window.

You will note that the if statement checks if the x value is greater than or equal to 10.5: this

is to ensure that it picks up the block, as it could have a value of 10.6. Remember to replace

the x, y, and z values with those from your photo booth. After the message is printed, the same

preview and camera snap will happen as before the while loop. The loop then resets itself so

you can enter it again and take another photo!

48 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 49

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

ch8listing1.py / Python 3

from mcpi.minecraft import Minecraft from picamera import PiCamera from time import sleep

mc = Minecraft.create() camera = PiCamera()

mc.postToChat("Find the photo booth")

camera.start_preview() sleep(2) camera.capture('/home/pi/selfie.jpg') camera.stop_preview()

while True:

x, y, z = mc.player.getPos()

sleep(3)

if x >= 10.5 and y == 9.0 and z == -44.3: mc.postToChat("You're in the photo booth!") sleep(1) mc.postToChat("Smile!") sleep(1) camera.start_preview() sleep(2) camera.capture('/home/pi/selfie.jpg') camera.stop_preview()

DOWNLOAD

magpi.cc/cameragit8

sleep(3)

49Chapter 8 Make a Minecraft photo booth

Page 50

Chapter 9

Make a spy camera

Set up a motion-activated spy camera in your room

e’ve all been there. You’ve gone

out for the day and you know you

come back and it’s slightly ajar. Who’s been in

there? Were they friend or foe? In this chapter

we’ll use the Camera Module or HQ Camera

as a spy camera that takes a picture when

anyone’s presence is detected by a passive

infrared (PIR) sensor. Here we’re using a

Raspberry Pi Zero – which is easier to hide away due to its size – with a special camera cable

for it, but you can use any Raspberry Pi model. Unless you want to power it from the mains,

you’ll also need a portable power supply such as a mobile phone battery pack.

using a Raspberry Pi Zero, you’ll need a special adapter cable since its camera connector is

smaller: the cable’s silver connectors should face the Raspberry Pi circuit board. You’ll also

need to have enabled the camera in Raspberry Pi Conguration, as explained in Chapter 1.

We’ll be using the picamera Python library to trigger our spy camera, so if you haven’t yet

installed it, open a Terminal window and enter:

sudo apt-get update sudo apt-get install python-picamera python3-picamera

closed your bedroom door, but you

Getting started

First, connect your Camera Module or HQ Camera to Raspberry Pi. Note that if you’re

YOU’LL NEED

• Camera Module / HQ Camera

• PIR sensor magpi.cc/pir

• Raspberry Pi Zero camera cable

(optional) magpi.cc/zerocamcable

• Portable power supply (optional)

• Jumper wires

Wire up the circuit

The circuit for this is fairly simple, especially as the PIR does not need a resistor as

part of its setup. The PIR comes with three connection pins: VCC, GND, and OUT. If you can’t

50 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 51

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Jumper wires are used to connect the

PIR’s pins to Raspberry Pi’s GPIO, either

to the pins of a header or to the hole

contacts if none has been attached

The passive infrared (PIR)

sensor will detect the

presence of anyone nearby

Figure 1 Connect a PIR sensor

nd their labels on the bottom of the sensor, lift off the plastic golf-ball-like diffuser and you

should see them on the top of the board. VCC needs to be connected to a 5V power pin,

GND needs to go to a ground pin, and then there’s the OUT wire which will be our input. We’re

connecting it to GPIO 14.

If your Raspberry Pi Zero has GPIO pins attached, you can use female-to-female jumper

wires to make the connections, as shown in Figure 1. Otherwise you can loop the wire around

the GPIO holes and use a bit of putty to keep them in place, or a dab of glue from a glue gun on

a low setting. Soldering is an option if you want to create a permanent spy camera device.

Write the code

Now we’ve got it all wired up, it’s time to start coding our spy camera. In the Thonny

code editor, create a new le and enter the code from ch9listing1.py. This script uses two

libraries: GPIO Zero and the standard picamera library. GPIO Zero can be used to get a reading

from the PIR motion sensor very easily, which can then be tied into the picamera code so it

takes a photo when motion is detected.

At the top, we import MotionSensor from GPIO Zero and PiCamera from picamera. Since

we’ll be giving each photo a timestamp, we also import datetime, along with sleep from the

time library.

51Chapter 9 Make a spy camera

Page 52

Camouaged against Yoshi, the camera will take candid snaps of

anyone who comes close to your gamecollection

In a never-ending while True: loop, we use GPIO Zero’s handy wait_for_motion function to pause the code until the PIR detects any motion. When it does, we set the photo le name to the current time and date, then take the picture. To enable the PIR to settle, we sleep for ve

seconds before returning to the top of the loop to wait for motion again.

Final preparations

You can run the code rst to give it a test. You might want to change the sensitivity

and/or trigger time, which you can do by adjusting the little orange potentiometer screws on

the side of the PIR board: Sx adjusts sensitivity, while Tx alters the trigger time.

Once that’s done, we’ll get the program to start automatically whenever we boot up the

Raspberry Pi. To do so, open up a Terminal window and edit the prole cong le with sudo nano /etc/profile. To the bottom of the le, add this line:

sudo python spy.py

In addition, to get Raspberry Pi to boot up slightly faster and, more importantly, to use a little

less power so your battery lasts longer, it’s best to get it to boot directly to the command line

rather than booting to the desktop. The easiest way to change this is to open Preferences >

Raspberry Pi Conguration from the desktop; in the default System tab, change Boot to the ‘To

52 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 53

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

ch9listing1.py / Python 3

#!/usr/bin/env python

from gpiozero import MotionSensor from picamera import PiCamera from datetime import datetime from time import sleep

sensor = MotionSensor(14) camera = PiCamera()

while True:

sensor.wait_for_motion() filename = datetime.now().strftime("%H.%M.%S_%Y-%m-

%d.jpg")

camera.capture(filename) sleep(5)

CLI’ option. Alternatively, open a Terminal window and enter sudo raspi-config to open the

Conguration Tool; select Boot Options > Desktop / CLI and option B2 – Console Autologin Text console.

Hide your camera

Now you need to nd a good place to hide your camera. The default cable for thecamera is limited by length, while the PIR can have its wires lengthened, so keep thatinmind when building your system. Alternatively, you could get a camera extender (magpi.cc/camextender) to link your cable to a standard-width one. Longer standard-width cables – of up to 2 m – are also available if you are not using a Raspberry Pi Zero.

Hiding your Raspberry Pi and battery behind a plush toy or photo frame can work well (you

could even put a dummy photo up and cut a hole in it for the camera to look through). The PIR

has quite a wide range, so put it up high where people are unlikely to look.

DOWNLOAD

magpi.cc/cameragit9

Check for intruders

All you need to do now is plug in the power supply and your Raspberry Pi Zero will turn

on and automatically run the script. Do some tests to make sure the camera is facing the right

way. Leave it running during the day and then when you get back, plug it into a monitor, stop

the script, and run startx to get the GUI up. From here you can see the pictures it has taken:

crucial evidence to catch your dog or sibling red-handed.

53Chapter 9 Make a spy camera

Page 54

Chapter 10

Smart door

Adding a Raspberry Pi to your door has magical results. Want

to see who’s at the door or know when the post has arrived?

Control the lock? Read on…

s your door a bore? Open and close, open and close. Snoozefest. Surely it can do more

than that? How about a smart door that knows when someone approaches, when the post

arrives, and can even offer remote viewing of the peephole? You can also add intelligent

lighting, a controllable door lock, and facial recognition, all powered with your Raspberry Pi. So,

let’s ignore super-expensive door systems and build our own. You can do as much, or as little,

as you like of this project and there’s plenty of room for new and inventive uses.

Prepare your Raspberry Pi

Although you can use any WiFi-capable model, this is a perfect project for the

Raspberry Pi3A+. Start by attaching Raspberry Pi to the Touch Display and preparing a microSD card with the latest Raspbian release. To allow easier access and mounting, we’ve

detached the control board from the back of the screen, taking great care of the ribbon cable.

Eventually, they’ll be put in a smart 3D-printed case. Now, get your Raspberry Pi set up and

make sure to sudo apt update && sudo apt upgrade before proceeding.

Attach the camera

We’re going to keep an eye on the outside world by replacing the door’s peephole

with the Raspberry Pi camera. A peep-hole is typically a two-piece barrel that screws together

YOU’LL NEED

• Raspberry Pi TouchDisplay magpi.cc/touch

• Camera Module / HQ Camera

• PIR sensor magpi.cc/pir

• 2 × Security door contact reed switch

magpi.cc/doorswitch

54 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

• Wired doorbell magpi.cc/wiredbell

• PAM8302 amplier magpi.cc/pam8302

• Speaker magpi.cc/3inspeaker

• Magnetic access control system

magpi.cc/magneticaccess

Page 55

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

and can be easily unscrewed from the inside.

Remove the barrel and cover the hole with

the camera. We’re just going to afx this with

tape for now; a printed mount will come later.

Mount the screen and Raspberry Pi to the door

(we used Command Strips), placed so you can

attach the camera’s ribbon cable to Raspberry

Pi once it’s shut down. Make sure the camera is enabled in Raspberry Pi Conguration.

Footsteps approaching!

The rst smart thing our door is going to do is detect someone approaching it. A cheap PIR sensor is perfect for the job. These cool little geodesic domes are triggered by heat and are the same gizmos that you nd in motion-sensor lights, switches, and security systems.

Connect to Raspberry Pi as shown in Figure 1, checking whether you have a 5 V or 3.3 V

sensor. Sensitivity and duration of a ‘detection’ can be controlled by the two potentiometers on

the PIR board. Mount this outside in a suitable location to ‘watch’ your door.

Monitor the door and letterbox

We have two magnetic reed switches, the type you nd on windows and doors for security systems. They are made up of two parts: the wired part is a reed switch and the

unwired a magnet. When the magnet meets the switch, it closes. If we attach the magnet

to the door and the switch to the frame, when the door opens, so does the switch. There’s

no polarity to worry about, so connect

one wire to GPIO 26 and the other to the adjacent GND. Repeat for the letterbox using GPIO 19. You may need

a breadboard.

NIGHT IS DARK

If you want the camera to work well at night,

you may want to consider a Pi NoIR Camera

Module supported with some infrared lighting.

Figure 1 The GPIO

wiring that’s needed for the various inputs and outputs

Ding dong!

If we replace the doorbell with

our own button, we can take a photo

with the Camera Module or HQ Camera

when someone presses it and send

a notication. Way better. Mount a

standard wired doorbell, which after all

is just a momentary contact button, to

the outside door frame and wire it back

to Raspberry Pi using GPIO 13 and an available GND pin. If you’re prototyping

on a breadboard, a tactile switch will

do ne.

55Chapter 10 Smart door

Page 56

The screen automatically

displays video when an

approaching person is detected

Cheap and simple reed switches

monitor the door and the letterbox

Sounds good

There’s little point in a doorbell that makes no sound. We can use the small, but surprisingly powerful, PAM8302 amplier with a speaker to make some noise. Supply power by soldering ‘Vin’ to an available 3V3 pin on Raspberry Pi, and ground to GND. To get an audio signal, you can tap the audio connector’s signal and ground, then connect them to A+ and Arespectively. Finally, solder the speaker to the larger + and - terminals. When prototyping, you

can skip this and use any active or passive speaker via the audio connector on Raspberry Pi.

Code

Double-check all your connections and then power up your Raspberry Pi. To use the code published here (overleaf), open a Terminal and enter:

mkdir ~/smartdoor nano ~/smartdoor/smartdoor_test.py

56 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 57

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Now type in the code as shown. Alternatively, to download all the code:

cd git clone https://github.com/themagpimag/cameraguideCh10

To enable it to play our doorbell sample:

sudo apt install mpg123

Now test with:

python3 ~/smartdoor/smartdoor_test.py

Watch the console output. If everything is working, you should be able to trigger the PIR, the

reed switches, and the doorbell. The camera will capture ten seconds of video when motion is detected, and a photo when the doorbell is pressed. These are both saved to the desktop.

Get alerts!

Let’s make this useful. Install Pushover on your phone, head over to pushover.net,

characters). Now create a new Application

and give it a name. Once created, you’ll

see an API Token. Make a note of this

too. From the GitHub repository, edit

smartdoor.py and add the User Key and

API Token where shown. Run this version

and you’ll get phone alerts for each event

and even a photo attachment when the

doorbell is pressed.

Intelligent

porch light

Following on from the Trådfri lighting

tutorial in The MagPi #75 (magpi.cc/75),

if you have an external porch light, why

not make it smart! The le porch.py will connect a Trådfri smart light to an API

that provides sunrise and sunset times

for your location. Leave the script running

and the light will switch on and off at the

correct times. Additionally, it monitors the

The web app can run on the touchscreen,

as well as on mobile devices or desktop browsers. Release the door fromanywhere!

57Chapter 10 Smart door

Page 58

You may want to prototype this project and test it before taking a drill to your door!

58 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 59

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

PIR sensor and will switch to full brightness

when someone approaches! To use the script,

get your latitude and longitude (you can use

Google Maps or Earth) and edit porch.py as

directed in the le.

Door lock

If you’re interested in being able to control your door’s lock, you may see that some

solutions are very pricey. One that is perfect for experimentation is the magnetic hold lock,

which uses an electromagnet to hold the door closed. The one we’ve used can withstand 180 kg of force, although stronger ones are available. The magnet mounts on the door and the electromagnet on the frame. The provided PSU contains a relay that can be powered by

Raspberry Pi by simply connecting it to a spare GPIO line and ground. Please note this is no

replacement for a proper door lock system.

Web app

If would be great to see what our door has been up to remotely, so a web app seems

the next logical step. In the directory called webapp is a Python script that uses Flask

to provide a web server that is usable on mobile devices. You can take a photo from the

peephole, see the last recorded video, and even control the magnetic door lock from Step 10.

Simply run the app alongside the others. Better still, set smartlights.py, porch.py, and webapp/smartdoor.py to start on boot (see the repository README).

GET THE RIGHT LOCK

Magnetic door locks vary in size and shape;

measure twice and order once!

Facial recognition

Once a futuristic technology, decent facial recognition is now well within the grasp

of Raspberry Pi. Using the doorbell photo taken by Raspberry Pi, we can recognise a face

using reference photos and send an alert to Pushover with the name of the caller! In a secure

environment, a recognised face could even trigger the lock or you could play a welcome

announcement. The install process is a little complicated, so if this interests you, see the

documentation in the GitHub repository in the face_recognition directory of the

‘smartdoor’ repository.

Over to you

Here we’ve given you the basics to get going, but more complex events are possible.

You could alert different people based on facial recognition or play custom doorbell tones.

And, if you had problems with deliveries, video evidence can build up automatically. On a

serious note, remember a lot of this is ‘just for fun’ and designed to inspire, so unless you’re

prepared to put in the work hardening the code and including failsafes, don’t rely on this, or

possibly make it as a fun kids’ door project (but maybe without the lock!).

59Chapter 10 Smart door

Page 60

smartdoor_test.py / Python 3

DOWNLOAD

from picamera import PiCamera from gpiozero import MotionSensor from gpiozero import Button from time import sleep import os import subprocess import sys

print('Getting smart...')

# Set up all our devices

camera = PiCamera() motion = MotionSensor(17) doorSensor = Button(26) letterbox = Button(19) doorbell = Button(13)

def motionDetected():

print('Motion detected, video recording') os.system('DISPLAY=:0 xset s reset') # Wakes the display up camera.start_preview() camera.start_recording('/home/pi/Desktop/motion.h264') sleep(10)

def motionStopped():

print('Stopping video recording') camera.stop_recording() camera.stop_preview()

magpi.cc/cameragit10

def doorOpen():

print('Door open')

def doorClosed():

print('Door closed')

def letterboxOpen():

print('You got mail!')

def doorbellPressed():

subprocess.Popen(['mpg123', '/home/pi/smartdoor/doorbell.mp3'], stdout=subprocess.PIPE, stderr=subprocess.STDOUT)

60 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 61

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

smartdoor_test.py (cont.) / Python 3

camera.capture('/home/pi/Desktop/doorbell.jpg') print('Someone\'s at the door!')

# Attach our functions to GPIOZero events

motion.when_motion = motionDetected motion.when_no_motion = motionStopped doorSensor.when_pressed = doorClosed doorSensor.when_released = doorOpen letterbox.when_released = letterboxOpen doorbell.when_released = doorbellPressed

print('Smart door is smart')

# Loop forever allowing events to do their thing

try:

while True: pass

except KeyboardInterrupt:

print('Smart door no longer smart')

except:

print('Oh dear')

61Chapter 10 Smart door

Page 62

Chapter 11

Car Spy Pi

Who’s that parked on the driveway?

Find out automatically using ANPR

utomatic number-plate recognition

(ANPR) is becoming more and

exclusive realm of the police, the technology

used to accurately read car number-plates

can now be found in supermarket and

airport car parks. It wasn’t long ago that

this technology was extremely expensive to

purchase and implement. Now, even a Raspberry Pi has the ability to read number-plates with

high accuracy using the Camera Module (or HQ Camera) and open-source software. Let’s see

what’s possible by building a system to detect and alert when a car comes onto the driveway.

applications, we’re going to see who’s home (or not) by reading number-plates of cars

coming and going on a driveway. This means our Raspberry Pi is probably going to live

outside; therefore, many environmental constraints come into place. You’ll need USB 5 V

power to your Raspberry Pi and a mounting position suitable for reading plates, although the

software is surprisingly tolerant of angles and heights, so don’t worry too much if you can’t

get it perfectly aligned.

you’ll need an appropriate enclosure. For a proper build, get an IP67 waterproof case (e.g.

magpi.cc/ip67kit). We’re opting for homemade, and are using a Raspberry Pi 3 A+ with the

RainBerry – a 3D-printable case that, once you add some rubber seals, provides adequate

protection. Make sure whatever you choose has a hole for the camera.

more commonplace. Once the

Pick a spot

First things rst: where are we going to put it? Although this project has lots of

Get an enclosure

As your Raspberry Pi is going to live outside (unless you have a well-placed window),

YOU’LL NEED

• Camera Module / HQ Camera

• Suitable outdoor enclosure,

e.g. magpi.cc/rainberry

• Pushover account (optional) pushover.net

62 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 63

The Camera Module is

protected from the elements by

a clear cover and rubber O-ring

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

This case is water-resistant with

rubber seals protecting the gaps

Prepare your Raspberry Pi

As we don’t need a graphical user interface, Raspbian Lite is our operating system

of choice. Any Raspberry Pi can handle this task, although if you want the fastest image

processing possible, you probably want to avoid the Zero models and get a nice, zippy Model

3B+ or 4. Get your operating system set up, make sure you’ve done the necessary sudo apt update && sudo apt -y upgrade and have congured WiFi if you’re not using Ethernet.

Finally, make sure your Camera Module (or HQ Camera) is connected and enabled. You can

check this by running sudo raspi-config and looking under ‘Interfacing Options’.

Install openALPR

Thankfully, we don’t need to be experts in machine learning and image processing to

implement ANPR. An open-source project, openALPR provides fast and accurate processing just from a camera image. ‘ALPR’ is not a mistake: this US project is ‘Automatic License Plate Recognition’. Thanks to APT, installation is straightforward. At the command line, enter the following:

sudo apt install openalpr openalpr-daemon openalpr-utils

libopenalpr-dev

This may take a while, as many supporting packages need to be installed, such as Tesseract,

an open-source optical character recognition (OCR) tool. This, coupled with code that

identies a number-plate, is what works the magic.

63Chapter 11 Car Spy Pi

Page 64

Protect your Raspberry Pi with a waterproof case

Time for a test

Once installed, you’ll be unceremoniously dropped back to the command prompt.

OpenALPR has installed a command-line tool to make testing its capabilities easier and they have also kindly provided a sample image. On the command line, enter the following:

cd wget http://plates.openalpr.com/ea7the.jpg

This is a sample USA plate image and a tough one too. Wget grabs the le from the web and places it in your home directory. Let’s see if we can recognise it:

alpr -c us ea7the.jpg

All being well, you’ll see a report on screen. The most likely ‘match’ should be the same as the le name: EA7THE.

Install Python libraries

We can use openALPR in Python, too. First, install the libraries using pip. If you don’t have pip installed, run:

64 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 65

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

sudo apt install python-pip

To install the libraries:

pip install openalpr picamera python-pushover

Now test everything is working by running Python (type python) and enter the following code

line by line at the >>> prompt:

import json from openalpr import Alpr

alpr = Alpr("us", "/etc/openalpr/openalpr.conf",

"/usr/share/openalpr/runtime_data")

results = alpr.recognize_file("/home/pi/ea7the.jpg") print(json.dumps(results, indent=4)) alpr.unload()

If you’ve not seen JSON-formatted text before, this might seem a bit much, but you should see

the correct plate number returned as the rst result.

Get a Pushover token

So that we can get an alert when a car arrives or leaves, we’re using old favourite

Pushover (pushover.net), which makes sending notications to mobile phones a breeze.

There’s a free trial, after which it’s a at fee of $4.99 per device, with no subscription or limits. Once logged in, go to ‘Your Applications’ and make a note of your User Key. Then click ‘Create an Application/API Token’. Call it ‘ANPR’, leave all the other elds blank, and click ‘Create Application’. Now make a note of the API Token; you’ll need this and the User Key for your code.

Typing time

Now you have everything you need

to create your ANPR application. Enter the

code listing shown here or download it from

magpi.cc/cameragit11. Save it as anpr.py in

your home directory. Edit the le and enter your User and App tokens where prompted. Save the le, then test by entering:

When a car arrives or leaves our driveway,

we receive an alert in seconds

65Chapter 11 Car Spy Pi

Page 66

python anpr.py

The code makes use of the Camera Module (or HQ Camera) and openALPR in tandem. Every ve seconds, the camera takes a picture which is passed to openALPR for analysis.

If a licence plate is found, we get the number. If there has been a change, an alert is sent to

Pushover, which is then forwarded to any registered mobile devices.

Make your list

If you want to have more friendly names rather than just the plate number, try adding a Python dictionary just after the import statements, like this:

lookup = { "ABC123": "Steve McQueen", "ZXY123": "Lewis Hamilton" }

Then change all instances of number_plate in the alert text as follows:

lookup[number_plate]

Now you’ll get a friendly name instead. See if you can handle what happens if the plate isn’t

recognised.

Run on boot

A key part of any ‘hands-free’ Raspberry Pi installation is ensuring that in the event of

a power failure, the required services start up again. There are many ways of doing this; we’re

going use one of the simpler methods.

sudo nano /etc/rc.local

Find the nal line, exit 0 and enter the following on the line above:

#Start ANPR Monitoring /usr/bin/python /home/pi/anpr.py

Press CTRL+X then Y to save the le. Finally, run the earlier pip command again, using sudo this time to install the libraries for the root user:

sudo pip install openalpr picamera python-pushover

On reboot, the code will start up and run in the background.

66 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 67

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Our target. The software does a great job of recognising number-plates

from different heights and angles

Add logging and curfews

One use of this installation is to track the times cars come and go. This can be

especially useful for young drivers who have curfew restrictions on their insurance. See if you

can augment the code to check whether a registration plate has not been seen after a certain

time. For example, if your younger family members have such a restriction, send them an alert

to their phone if their car isn’t in the driveway 30 minutes beforehand. You might save them an

insurance premium increase! Also, why not log all the comings and goings to a le? A bit of

data analysis might help reduce car usage or fuel costs.

Make it your own

As ever, it’s over to you. Now you have the ability to track and record registration

plates, there are many different applications for you to explore. Since all the analysis of the

image is done locally, no internet connection is required for the system to work. Is there

someone ‘borrowing’ your parking space at work? Catch ’em in the act! Why not take your CarSpy Pi on the road? It could record every vehicle you encounter, which may be useful should something untoward happen. Combine a Raspberry Pi Zero with a ZeroView

(magpi.cc/zeroview) and you’re all set.

67Chapter 11 Car Spy Pi

Page 68

anpr.py / Python 3

DOWNLOAD

from openalpr import Alpr from picamera import PiCamera from time import sleep import pushover

# Pushover settings

PUSHOVER_USER_KEY = "<REPLACE WITH USER KEY>" PUSHOVER_APP_TOKEN = "<REPLACE WITH APP TOKEN>"

# 'gb' means we want to recognise UK plates, many others are available

alpr = Alpr("gb", "/etc/openalpr/openalpr.conf", "/usr/share/openalpr/runtime_data") camera = PiCamera() pushover.init(PUSHOVER_APP_TOKEN) last_seen = None

try:

# Let's loop forever: while True:

# Take a photo print('Taking a photo') camera.capture('/home/pi/latest.jpg')

magpi.cc/cameragit11

# Ask OpenALPR what it thinks analysis = alpr.recognize_file("/home/pi/latest.jpg")

# If no results, no car! if len(analysis['results']) == 0: print('No number plate detected')

# Has a car left? if last_seen is not None: pushover.Client(PUSHOVER_USER_KEY).send_message( last_seen + " left", title="Driveway")

last_seen = None

else:

68 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 69

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

anpr.py (cont.) / Python 3

number_plate = analysis['results'][0]['plate'] print('Number plate detected: ' + number_plate)

# Has there been a change? if last_seen is None: pushover.Client(PUSHOVER_USER_KEY).send_message( number_plate + " has arrived",

title="Driveway")

elif number_plate != last_seen: pushover.Client(PUSHOVER_USER_KEY).send_message( number_plate + " arrived and " + last_seen +

" left",

title="Driveway")

last_seen = number_plate

# Wait for five seconds sleep(5)

except KeyboardInterrupt:

print('Shutting down') alpr.unload()

69Chapter 11 Car Spy Pi

Page 70

Chapter 12

Build a wildlife camera trap

Uncover the goings-on in your garden, pond, or school

playground when no one’s looking with this easy-to-use

Raspberry Pi camera trap

ver wondered what lurks at the bottom of your garden at night, or which furry

friends are visiting the school playground once all the children have gone home?

Using a Raspberry Pi and Camera Module (or HQ Camera), along with Google’s

Vision API, is a cheap but effective way to capture some excellent close-ups of foxes, birds,

mice, squirrels and badgers, and to tweet the results.

Using Google’s Vision API makes it really easy to get AI to classify our own images. We’ll

install and set up some motion detection, link to our Vision API, and then tweet the picture if

there’s a bird in it. It’s assumed you are using a new Raspbian installation on your Raspberry

Pi and you have your Raspberry Pi camera set up (see Chapter 1). You will also need a Twitter

account and a Google account to set up the APIs.

Motion detection with Pi-timolo

There are many different motion-detection libraries available, but Pi-timolo was chosen as it is

easy to edit the Python source code.

YOU’LL NEED

• Camera Module / HQ Camera

• Pi NoIR Camera Module (optional)

magpi.cc/ircamera

• Waterproof container (like a jam jar)

70 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

• ZeroCam NightVision (optional)

magpi.cc/zerocamnight

• Blu Tack, Sugru, elastic bands, carabiners

• ZeroView (optional) magpi.cc/zeroview

Page 71

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Position the camera low

and ensure it’s tight up

against any glass or plastic

Get a decent power bank that will last for

at least seven or eight hours overnight

Use sealable containers and jars

and make sure they’re watertight

First, make sure Raspbian is up to date. In a Terminal, enter:

sudo apt update && sudo apt upgrade

Then use the following command to perform an automatic installation of Pi-timolo:

curl -L https://raw.github.com/pageauc/pi-timolo/master/source/

pi-timolo-install.sh | bash

Once installed, test it by typing in cd ~/pi-timolo and then ./pi-timolo.py to run the

Python script. At this point, you should be alerted to any errors such as the camera not being

installed correctly, otherwise the script will run and you should see debug info in the Terminal

window. Check the pictures by waving your hand in front of the camera, then looking in

Pi-timolo > Media Recent > Motion. You may need to change the image size and orientation of the camera; in the Terminal window, enter nano config.py andedit these variables:

imageWidth, imageHeight, and imageRotation.

While we’re here, if you get a lot of false positives, try changing the motionTrackMinArea

and motionTrackTrigLen variables and experiment with the values by increasing to reduce

sensitivity. See the Pi-timolo GitHub repo (magpi.cc/pitimologit) for more details.

71Chapter 12 Build a wildlife camera trap

Page 72

Unleash your inner

There will also be some editing of the pi-timolo.py le, so don’t close the Terminal

window. Code needs to be added to import some Python libraries (ch12listing1.py), and

tweeting (ch12listing2.py). We can do this now in preparation of setting up our Google and

Twitter API. You should still be in the Pi-timolo folder, so type nano pi-timolo.py and add

UserMotionCodeHere() function and where it’s called from. Add the new code into the function

(line 240), before the return line. Also locate where the function is being called from (line 1798),

up the APIs.

Springwatch

Animal detection and tweeting

We will be using Google Label Detection, which returns a list it associates with the image. First

off, you will need to install the Google Cloud Vision libraries on your Raspberry Pi, so type pip

install --upgrade google-cloud-vision into your Terminal window. Once nished, run

pip install google-cloud-storage.

a new project (you may need to log in or create a Google account). Go to the API Dashboard

and search for and enable the Cloud Vision API. Then go to API & Services > Credentials,

click on Create Credentials and select Service Account from the drop-down. Fill in the details,

then click Create. Don’t bother to select a Role, just click Continue. Click Create Key and you’ll

72 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 73

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

pi-timolo folder and make a note of the le path. Next, go back to pi-timolo.py and add the

line: os.environ["GOOGLE_APPLICATION_CREDENTIALS"] = "path_to_your_.json_ credential_file" below import os to reference the credentials in your JSON le.

Finally, set up a Twitter account if you haven’t already and install Tweepy by entering sudo

pip install tweepy into a Terminal window. Once set up, visit apps.twitter.com and create

a new app, then click on Keys and Access Tokens. Edit the code in userMotionCodeHere()

with your own consumer and access info, labelled as ‘XXX’ in the code listing. Finally, place

your camera in front of your bird feeder and run ./pi-timolo.py. Any pictures taken of a bird

should now be tweeted! If you want to identify a different animal, change the line if "bird"

in tweetText: animalInPic = true.

Please note that although the API works well, it can’t always discern exactly what’s in

the picture, especially if partially in view. It also won’t distinguish between types of bird, but

you should have more success with mammals. You can test the API out with some of your

pictures at magpi.cc/visionai and visit twitter.com/pibirdbrain to see example tweets (scroll

down a bit). Good luck and happy tweeting!

Get great photos with night-vision cameras

73Chapter 12 Build a wildlife camera trap

Page 74

ch12listing1.py / Python

DOWNLOAD

# add this in at the very top, under

print('Loading ....') along with the other

libraries imported

import io import tweepy from google.cloud import vision from google.cloud.vision import types from google.cloud import storage

magpi.cc/cameragit12

ch12listing2.py / Python

# search for userMotionCodeHere. # There will be 2 results, # edit the second so you are passing filename to the function

userMotionCodeHere(filename)

# make sure you include filename as a parameter in the function

def userMotionCodeHere(filename):

# we need to create an instance of the Google Vision API client = storage.Client() # instantiates a client client = vision.ImageAnnotatorClient()

# loads the image into memory with io.open(filename, 'rb') as image_file: content = image_file.read()

image = types.Image(content=content)

# performs label detection on the image file response = client.label_detection(image=image) # pass the response into a variable labels = response.label_annotations

# we have our labels, now create a string to add to the text # for debugging - let’s see what Google thinks is in the image print('Labels:') # add labels to our tweet text tweetText = "Labels: "

74 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 75

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

ch12listing2.py (cont.) / Python

animalInPic = False for label in labels: print(label.description) tweetText = tweetText + " " + label.description # edit this line to change the animal you want to detect if "bird" in tweetText: animalInPic = true

# set up Tweepy

# consumer keys and access tokens, used for authorisation

consumer_key = ‘XXX’ consumer_secret = ‘XXX’ access_token = ‘XXX’ access_token_secret = ‘XXX’

# authorisation process, using the keys and tokens auth = tweepy.OAuthHandler(consumer_key, consumer_secret) auth.set_access_token(access_token, access_token_secret)

# creation of the actual interface, using authentication api = tweepy.API(auth)

# send the tweet with photo and message photo_path = filename # only send tweet if it contains a desired animal if animalInPic: api.update_with_media(photo_path, status=tweetText)

return

75Chapter 12 Build a wildlife camera trap

Page 76

Chapter 13

Take your camera underwater

Explore the underwater world with your camera

here are plenty of underwater sports cameras available, but they can be quite

expensive, especially if you want to be able to control them remotely. In this

customisable camera unit. There are lots of options and alternative sources of components

for a project like this. For example, the Pimoroni Enviro board (or earlier Enviro pHAT) can

report back information about the environment in which the camera is operating, especially

how much light is available.

chapter we’re going to use readily available Raspberry Pi add-ons to make a cheaper,

Find a suitable container

To protect the electronics inside it, the container for your Raspberry Pi and camera

needs to be watertight and to have at least a see-through lid. You can nd food container

boxes with a very tight seal, but these tend to be translucent rather than transparent. The

size of box will probably determine your choice of Raspberry Pi model andpower source.

Raspberry Pi Zero W boards are great as they are so small and have built-in wireless LAN.

You can also save space by using a LiPo battery instead of a standardpower bank (although you’ll need a boost regulator too, such as the Pimoroni ZeroLiPo).

YOU’LL NEED

• Camera Module / HQ Camera

• Transparent, waterproof box

magpi.cc/waterproofcase

• Portable power source

• hostapd and dnsmasq packages

76 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

• Python Flask library

• WiFi dongle (if not using a Raspberry Pi model

with built-in wireless LAN)

• Enviro board (optional) magpi.cc/enviro

• ZeroView (optional) magpi.cc/zeroview

Page 77

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

You can save space by using a LiPo battery (via a boost regulator) instead of a power bank

Set Raspberry Pi as a wireless access point

Start from a fresh Raspbian install on a microSD card. Open up a Terminal window and

enter the following commands to update the APT database and install the required packages:

sudo apt-get update sudo apt-get install -y dnsmasq hostapd python3-flask

First, since the conguration les aren’t ready yet, turn off dnsmasq and hostapd:

sudo systemctl stop dnsmasq sudo systemctl stop hostapd

Next, congure your wireless interface to have a static IP address:

sudo nano /etc/dhcpcd.conf

Go to the end of the le and edit it so that it looks like the following (we're using the IP address

192.168.4.1, but you may want to choose a different one):

interface wlan0 static ip_address=192.168.4.1/24 nohook wpa_supplicant

77Chapter 13 Take your camera underwater

Page 78

You’ll still have to get pretty close to the water yourself

Now restart the dhcpcd daemon and set up the new wlan0 conguration:

sudo service dhcpcd restart

Next, you need to edit the /etc/dnsmasq.conf and /etc/hostapd/hostapd.conf les – see magpi.cc/accesspoint for details – ensuring that the IP addresses are consistent with your

settings in /etc/dhcpcd.conf. Then reboot your Raspberry Pi.

Add the Enviro board

Pimoroni’s Enviro board (or the earlier Enviro pHAT, which is still available from some

online retailers) enables you to send back environmental data from the camera. With your

Raspberry Pi powered off, mount the board on its GPIO header.

Now power up Raspberry Pi and install the Python library and dependencies using the

following Terminal commands:

git clone https://github.com/pimoroni/enviroplus-python cd enviroplus-python sudo ./install.sh sudo pip3 install smbus2

78 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 79

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

The Enviro board’s library comes with some example programs; you should run some of these

to test that everything is working correctly.

Note: If you are not using an Enviro board, you willneed to comment out some of the related

code in the main ch13listing1.py script. IfusingtheEnviro pHAT instead, use the alternative

ch13listing2.py script in the GitHub repository (magpi.cc/cameragit13).

Fitting everything into your container

To cut down on reections and obtain the best possible images, the camera should be

as close to the transparent side of your container as possible. The ZeroView from the Pi Hut

is a clever mounting plate that uses suction cups and will also hold your Raspberry Pi Zero

securely. Alternatively, you could make a mount out of cardboard and glue this to the inside of

the container. Velcro tape can be a good solution for power sources (which normally need to

be removable for recharging).

Add some code, HTML and CSS

Clone the project’s GitHub repository onto Raspberry Pi. In a Terminal window, enter:

git clone https://github.com/themagpimag/cameraguideCh13

Then use the desktop File Manager to move the Flask folder within cameraCh13 to your

Raspberry Pi’s home directory (or use the mv command in a Terminal window).

A makeshift handle to lower the waterproof box into the water

79Chapter 13 Take your camera underwater

Page 80

To run the program, enter the following Terminal command:

sudo python3 /home/pi/Flask/ch13listing1.py

To see the generated webpage from another computer, you just have to open a web browser

and enter your Raspberry Pi’s static IP address. Using the on-screen buttons, we can also

switch between recording modes (video or continuous still frames) or take photos on demand

– by selecting QuickSnap and then clicking the Take button. This control of the camera is achieved via the picamera library, which is used for the three main functions – timelapse,

video, and snapstart – dened in our Python script. You could enhance the project by adding

additional exposure and shutter speed controls to your interface if you want.

Note: To see the latest image taken, press an on-screen button or reload the webpage.

Set the code to run at boot

Naturally, we’ll want the code to run automatically whenever the Raspberry Pi boots up. To do so, add this line to your /etc/rc.local le, immediately above the exit 0 line:

sudo python3 /home/pi/Flask/ch13listing1.py &

It is also a good idea to congure your

Raspberry Pi to only boot to the command

line rather than the desktop, as this uses a

little less power and prolongs battery life.

The easiest way to change this is to open

Preferences > Raspberry Pi Conguration

from the desktop; in the default System

tab, change Boot to the ‘To CLI’ option.

Alternatively, open a Terminal window and

enter sudo raspi-config to open the

Conguration Tool; select Boot Options > Desktop / CLI and option B2 – Console

Autologin Text console.

Now go and nd somewhere wet! You

might want to run a few tests in the bath

before venturing further aeld.

The web interface shows

environmental information and lets you control the camera

80 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 81

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

ch13listing1.py / Python 3

from flask import Flask, render_template,

request, redirect, url_for

import time, os, shutil from picamera import PiCamera from datetime import datetime, timedelta from threading import Thread from ltr559 import LTR559 from bme280 import BME280 from smbus2 import SMBus

ltr559 = LTR559() bus = SMBus(1) bme280 = BME280(i2c_dev=bus)

app = Flask(__name__) app.config['SEND_FILE_MAX_AGE_DEFAULT'] = 1

def timelapse(): # continuous shooting

cam = PiCamera() cam.resolution = (1640,922)

for filename in cam.capture_continuous('img{timestamp:

%H%M%S

print('snap taken') print(btn1,btn2) shutil.copyfile(filename,'/home/pi/Flask/static/latest.

jpg')

if btn1 != 's': break cam.close() print('timelapse thread stopped')

}.jpg'):

DOWNLOAD

magpi.cc/cameragit13

%Y%m%d

def video(): # record a video

cam = PiCamera() t='{:%Y%m%d-%H%M%S}'.format(datetime.now()) cam.resolution = (1920,1080) cam.start_recording('vid'+t+'.h264') while btn1 == 'v': print(btn1,btn2) pass cam.stop_recording() cam.close() print('video thread stopped')

81Chapter 13 Take your camera underwater

Page 82

ch13listing1.py (cont.) / Python 3

def snapstart(): # take pictures on demand

cam = PiCamera() cam.resolution = (1640,922) print('entered snapshot mode') global btn2 while btn1 == 'q': time.sleep(0.1) if btn2 == 'a': print('taken snap: btn2 =' + btn2) t='{:%Y%m%d-%H%M%S}'.format(datetime.now()) filename = 'snap'+t+'.jpg' cam.capture(filename) shutil.copyfile(filename,'/home/pi/Flask/static/

latest.jpg')

btn2 = 'o' print('btn2 =' + btn2)

cam.close() print('exiting snaphot mode')

# we are able to make two different requests on our webpage # GET = we just type in the url # POST = some sort of form submission like a button

@app.route('/', methods = ['POST','GET'])

def hello_world():

status = 'off' global btn1 btn1 = 'o' global btn2 btn2 = 'o' message = 'All good '

# if we make a post request on the webpage aka press button

then do stuff

if request.method == 'POST':

# if we press the turn on button if request.form['submit'] == 'Video':

82 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 83

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

ch13listing1.py (cont.) / Python 3

print('BP: Recording video') status = 'video' btn1 = 'v' t2 = Thread(target=video) t2.start() message = 'All good' elif request.form['submit'] == 'Video Off': print('BP: Video off') status = 'Idle' btn1 = 'o' message = 'All good' elif request.form['submit'] == 'Stills': print('BP: Recording stills') btn1 = 's' t1 = Thread(target=timelapse) t1.start() status = 'stills' message = 'All good' elif request.form['submit'] == 'Stills Off': print('BP: stills off') status = 'Idle' btn1 = 'o' message = 'All good' elif request.form['submit'] == 'QuickSnap': print('BP: QuickSnap') status = 'Ready to snap' btn1 = 'q' t3 = Thread(target=snapstart) t3.start() message = 'All good' elif request.form['submit'] == 'QuickSnap Off': print('BP:QuickSnap off') status = 'Idle' btn1 = 'o' message = 'All good' elif request.form['submit'] == 'Take': print('BP:Take') status = 'Snapshot mode' btn1 = 'q' btn2 = 'a' message = 'All good' elif request.form['submit'] == '_Take_':

83Chapter 13 Take your camera underwater

Page 84

ch13listing1.py (cont.) / Python 3

print('BP:Take error') status = 'Error' message = 'Enable QuickSnap first' btn1 = 'o' else: pass

temp = round(bme280.get_temperature(),2) # temperature press = int(bme280.get_pressure()) # pressure lux = ltr559.get_lux() # light levels df = os.statvfs('/') # check if we're running out of disk

space

df_size = df.f_frsize * df.f_blocks df_avail = df.f_frsize * df.f_bfree df_pc = round(100 -(100 * df_avail/df_size),1) print(btn1, btn2)

# the default page to display will be our template with our

template variables

return render_template('index2.html', message= message,

status=status, temp=temp, press=press, lux=lux, df_pc=df_pc, btn1 = btn1)

if __name__ == "__main__":

# let's launch our webpage!

# do 0.0.0.0 so that we can log into this webpage # using another computer on the same network later # specify port 80 rather than default 5000

app.run(host='0.0.0.0',port=80,debug=True)

84 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 85

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

85Chapter 13 Take your camera underwater

Page 86

Chapter 14

Install a bird box camera

Observe nesting birds without disturbing them

hile it’s simple enough to set up a

Camera Module in a weatherproof

garden, for this project we’ll be installing a

camera inside a bird box. Since it’ll be dark

inside, and we can’t use a standard light

source, we’ll need to use a Pi NoIR Camera

Module. ‘NoIR’ stands for ‘no infrared’, as

it omits the IR lter found in the standard

camera. This enables you to use an infrared

light source to see in the dark. Note that we’ll

need to adjust the xed focus of the camera by

unscrewing the lens.

insects and predators, and so would deter any birds from nesting there. So we need to use a Pi NoIR Camera Module (Figure 1). Apart form the omission of an infrared lter, this works

exactly the same way as the standard camera, so you can connect it up to your Raspberry Pi

as in Chapter 1 and use all the same Terminal commands. So, for instance, you can obtain a

video preview with:

raspivid -t 0

box to observe wildlife in your

Set up the Pi NoIR

We can’t use a standard light source inside the bird box, since this could attract

YOU’LL NEED

• Pi NoIR Camera Module

• Focus adjustment tool

• Bird box

• IR LED

• Female-to-female jumper wires

• Power source

• Ethernet cable (optional)

You’ll notice that everything looks a little strange; this is because you’re looking at a

combination of visible light and infrared light. To test it out in darkness, turn the lights off, aim

86 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 87

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

a TV remote control at your face and press

the buttons to produce an IR light source.

You should see your face illuminated in the

darkness. The image will look black and white

(greyscale), because there are no wavelengths

of light from the visible spectrum being

detected. However, a black and white image

is good enough to allow you to watch what’s

happening inside a bird box, and it doesn’t disturb or interfere with the birds in any way. Press CTRL+C to exit the preview.

Wire up an IR LED

We’ll need a suitable infrared light source in the bird box. In this example we’re using

a single IR LED, but alternatives include small IR lamps and the IR version of the LISIPAROI (lisiparoi.com). Our 890 nm IR LED is an identical component to the ones found inside TV

remote controls; the only difference is that we’re going to keep it on constantly when shooting

video or stills in the bird box.

As usual, you should turn off your Raspberry Pi before connecting anything up. If you’ve

wired up an LED to Raspberry Pi’s GPIO pins before, then please note that this LED needs to

be done slightly differently. Since an infrared LED requires more current than the GPIO pins

can provide, it needs to be connected directly to the 5V supply of Raspberry Pi with a 220 Ω

resistor inline; without the resistor the current will be too high, and the LED will burn out after

about ten seconds.

HQ CAMERA

If you want to use an HQ Camera for this

project, you’ll need to remove its IR lter so

it can capture infrared images in the dark.

See the leaet supplied with the camera

for more details, but note that this is an

irreversible process.

Figure 1 The

PiNoIR Camera

Module can see in the dark with infrared lighting

87Chapter 14 Install a bird box camera

Page 88

The longest leg of the IR LED is

the anode – connect it to a 5 V pin

The 222 Ω resistor limits the

Figure 2

Figure 2 shows how the LED should be wired up. You’ll notice that the LED has two legs, one

slightly longer than the other. The longer of the two is called the anode and the shorter is the

cathode. The LED needs power to ow into the anode and out of the cathode; if you get the

polarity wrong then nothing will happen.

Use a couple of female-to-female jumper wires to make the following connections. Connect

the anode (long leg) to 5 V, which is the rst pin on the outside row on the GPIO. Connect the cathode (short leg) to the 220 Ω resistor. Connect the other side of the resistor to ground

(GND), which is the third pin in on the outside row of the GPIO.

current owing through the LED

Test the LED

With everything wired up correctly, turn your Raspberry Pi back on. You’ll notice that

the infrared LED doesn’t appear to be working, but in fact it is. Your human eyes can’t see it,

but the Pi NoIR camera can. Turn on the camera preview again with raspivid -t 0. Hold the

LED in front of the camera and you should see it lit. If not, then you may have mixed up the

polarity of the anode and cathode. Double-check your wiring against the Figure 2 diagram.

Try turning out the lights and aiming the LED at yourself; don’t look directly into it, however, as

infrared light can still cause harm to your eyes. You’ll see from the Pi NoIR camera preview that it will illuminate you quite well. Press CTRL+C when you want to exit.

Adjust focus

By default, the Pi NoIR Camera Module has a xed focal length of 50 cm and depth of eld of 50 cm to innity. This means that objects will only appear in focus if they’re at least 50 cm away from the lens of the camera. The Gardman box we’re using in this example has an interior height of 18 cm, so we’ll denitely need to shorten the focal length.

88 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 89

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Like the standard Camera Module,

the Pi NoIR has a lens that can rotate to

adjust the focus. If your camera came

supplied with a focus adjustment tool (a

little plastic wheel), you can use this to turn

the lens, removing three blobs of silicone

in the process. If you don’t have the tool,

you could always 3D-print one; there are

numerous design available on Thingiverse,

e.g. thingiverse.com/thing:2533691.

Alternatively, you could always use a scalpel

Figure 3 Pinch the lid and then use a pen

to mark a cross where your thumb is

Carefully rotate it anticlockwise a few times – be careful not to rotate the lens too far,

otherwise it will pop out, and it can be a bit tricky to get it back in and on the thread. If this

does happen, just put it back in gently and rotate clockwise until it catches.

Now reconnect the camera to Raspberry Pi. Place an object – such as a watch or business

card – in the bottom of the bird box, then remove its roof (remove the screw), hold the camera at the approximate height of the roof, and look at the camera preview. You may wish to put

something under the object at this point to simulate the height of a nest, to make doubly sure

that the birds will be in focus. Remember that once birds move in, you can’t come back and

adjust the camera if the focus is wrong.

to carefully remove the silicone blobs and

then use tweezers to rotate the lens – see

magpi.cc/adjustfocus for more details.

Install the camera

Place your nger on the roof, approximately above the centre of the main body of

the bird box. Lift up the roof and place your thumb directly below your nger, so that you’re

pinching the lid as shown in Figure 3. Your thumb is now where the camera needs to be.

Take a pen and mark this spot with a cross. Cut out a rectangle of cardboard approximately 4 cm× 2 cm (1.5" × 0.75") and fold it in half

lengthways. Use some tape to secure it to

the underside of the roof so that it’s a few

millimetres below the cross. This is going to

be used to compensate for the angle of the

roof, so that the camera points directly into

the middle of the bird box.

Next, take the Pi NoIR and slide the exible cable down between the roof hinge and the back wall of the box – Figure 4. Do

this with the tin connectors facing away

from the back wall. If you wish, you can

Figure 4 Slide the camera cable between

the roof hinge and the back wall of the box

89Chapter 14 Install a bird box camera

Page 90

remove the two middle staples holding the

hinge in place; this will make the ex exit the bird box more neatly.

Take some tape and put it across the

top of the Pi NoIR board – do not cover the

camera lens! Secure the camera in place

so that the central lens is directly over the

cross that you drew earlier. The camera

should sit at an angle – Figure 5. Close the

lid and inspect the camera angle from the

Figure 5 When in place, the camera should sit

at an angle to compensate for the roof slope

alternative to taping it in place would be to use the four mounting holes to screw it to the lid

via a wedge of wood instead of the cardboard.

Add the LED

Secure the infrared LED to the underside of the roof. Don’t attach it too close to the

camera, or you’ll see a lot of glare on the video. The LED can go anywhere, but it can help to

bend its legs by 90 degrees, as shown in Figure 6, and secure it to the roof that way. You may

also wish to blank off the end of the LED with correction uid or by ling it down with a nail le.

This will prevent any spotlight effect on the video and give a more diffuse light.

side: it needs to point directly at the centre

of the base. If it doesn’t look right, then go

back and adjust it until you’re happy. An

Test it again

Now reconnect your Raspberry Pi and test the focus once again. We recommend

connecting the camera ex coming from the back of the bird box to Raspberry Pi rst. Then connect the LED and resistor, followed by the screen, keyboard, and nally the power supply.

When testing this setup, it can be helpful to rest Raspberry Pi upside-down on the roof of the bird

box, but do whatever works best for you.

Boot up as usual and then start the video

preview with raspivid -t 0. With the

roof of the bird box closed, you should be

able to see the inside in black and white.

This shows that the infrared illumination is

working; you should even be able to cover

the hole and still see the inside. It will look

similar to Figure 7, but will be slightly more

zoomed in. This is because this image was

taken using the raspistill command and

not raspivid. If you can’t see anything

at all, then it’s likely the LED is not wired

90 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Figure 6 The IR LED is taped to the underside

of the roof, not too close to the camera

Page 91

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

up correctly: double-check the wiring and the

polarity of the anode and cathode.

It’s now helpful to put an object with some

black-on-white text into the bird box to verify

the focus, such as a watch or business card.

Ensure that the text is in focus and readable;

adjust the camera focus again as necessary

before continuing. Remember to compensate

for the nest height. Press CTRL+C when you

want to stop the camera preview.

Weatherproof it

While you can attach your Raspberry Pi directly to the outside of the bird box, an

alternative is to use a longer camera cable. Either way, you’ll need to put Raspberry Pi inside a

weatherproof box. Preventing water getting into the bird box should also be a priority. The roof

could be sealed using silicone sealant, which is often used to seal the edges of windows and

bathroom sinks. Choosing a site which is beneath the overhang of an existing roof will help a lot, so the bird box will not be rained on directly.

Lastly, you need to consider how you will get power and an internet connection to the bird

box? You could use a wireless USB dongle, or the built-in wireless LAN of a Raspberry Pi 3 / 3B+ / 4 / Zero W, but Ethernet is more reliable for streaming video, especially in built-up areas that have a lot of wireless trafc.

TURN OFF RED LED

If you are using an original (Rev 1.3) Pi NoIR

Camera Module, you’ll need to disable its

red LED that lights up whenever the camera

is on. Enter sudo nano /boot/cong.txt

and add the following line to the end of the

le: disable_camera_led=1. Save and exit

the le, then reboot.

Obtain images

With everything installed, connected, and powered up, you can SSH in to your

Raspberry Pi from another computer (see magpi.cc/ssh for details) to control it remotely.

You are then able to enter standard Terminal commands such as raspistill and raspivid

to obtain stills (including time-lapses – see

Chapter 3) and video footage. You could

also write one or more Python scripts

using the picamera library.

Note that you can’t view the live camera

preview via SSH. However, you are able to

live-stream video from the bird box. This

could be achieved using a client-server

setup, as described in Chapter 15, to pipe

the output to a video player on the client

computer. Alternatively, you could make

use of an internet video service offering

Figure 7 Make sure that the test object is

raised up slightly and the text is in focus

livestreaming, such as YouTube (see

magpi.cc/birdboxyt for details).

91Chapter 14 Install a bird box camera

Page 92

Chapter 15

Live-stream video and stills

Stream video and regular stills to a remote computer

ne of the drawbacks of using SSH

or VNC to access your Camera

Raspberry Pi remotely from another

computer is that you can’t (typically) view

the camera preview via these methods. To

get around this, you’ll need to stream live video across the network. While there are various

methods available for doing this, in this chapter we’ll show you how to create a client-server

setup for video streaming using the picamera Python library. We’ll also explore how to send

a stream of stills over the network.

there is an easier method, explained in Step 03.

read the video stream (which we’ve yet to write to the code to create) and pipe it to a media

player. Note that while you can use a Raspberry Pi 4 for the task of remote playback, earlier

Raspberry Pi models won't work in this role since the CPU is not powerful enough to do the

video decoding (and neither VLC nor MPlayer supports doing this using the GPU). Therefore

you may need to run this script on a faster machine, although even an Atom-powered

netbook should be quick enough for the task at non-HD resolutions.

connections on 0.0.0.0:8000, i.e. all IP addresses on the local network. We then accept a

single connection and make a le-like object out of it.

use MPlayer instead of VLC, add a # to the start of the cmdline = ['vlc… line to comment

it out, and remove the # from the cmdline = ['mplayer… line.

Module- or HQ Camera-equipped

Server-side script

Note: If you are using a Linux-based computer for playback of the video stream,

First, we’ll write a Python server script, ch15listing1.py, for the remote computer that will

After importing the libraries required at the top of the script, we start listening for

In the try: block, we run a media player from the command line to view it – if you want to

YOU’LL NEED

• Camera Module / HQ Camera

• Remote computer

92 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 93

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

The remote server script reads the video stream and pipes it to a media player, such as VLC

In the while True: loop, we repeatedly read 1kB of data from the connection and write it to

the selected media player’s stdin (standard input) to display it.

Note: If you run this script on Windows or macOS, you will probably need to provide a

complete path to the VLC or MPlayer executable/app. If you run the script on macOS, and

are using Python installed from MacPorts, please ensure you have also installed VLC or

MPlayer from MacPorts.

Client-side script

Now we’ll create a client script, ch15listing2.py, on our Raspberry Pi with the

Camera Module or HQ Camera equipped. This will connect to the network socket of our

server (playback) script to send a video stream to it.

After importing the required libraries at the top, we connect a client socket to

my_server:8000 – you’ll need to change my_server to the host name of your server

(thecomputer that will playing back the stream). If you are using a Linux PC or Mac, just type hostname -I in a Terminal window to nd it out; in Windows, it’s the Computer Name inControl Panel > System.

We then create a le-like object from the network socket before triggering the camera

to start recording. In this example we’re using a resolution of 640×480 with a frame rate of

24 fps, but you can adjust these numbers to your requirements. We’ve also set the camera

to record for 60 seconds with camera.wait_recording(60); again, you can change this

number to suit your preference.

93Chapter 15 Live-stream video and stills

Page 94

An alternative is to set Raspberry Pi as the server and open the

network source in VLC or another media player

Run the server script, then the client script. You should see the video stream played in

your chosen media player. You may notice some latency; this is normal and due to buffering

by the media player.

Server-side script

As mentioned, if you’re using a Linux PC for playback of the video stream, there is a

much quicker and easier way to achieve what we’ve done in Steps 01 and 02. On the server

(playback) machine, enter the following command into a Terminal window:

nc -l 8000 | vlc --demux h264 -

Then, on the client – your Raspberry Pi with the Camera Module or HQ Camera – issue the

following command:

raspivid -w 640 -h 480 -t 60000 -o - | nc my_server 8000

…replacing my_server with the server’s host name.

Switch it around

An alternative method is to reverse the direction so that Raspberry Pi acts as a

server. We can then get it to wait for a connection from the client before streaming video.

Enter the ch15listing3.py example on Raspberry Pi and run it.

94 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 95

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

The big advantage of this method is that you then only need to use a single command to

initiate playback on the remote computer:

vlc tcp/h264://my_pi_address:8000/

…replacing my_pi_address with your Raspberry Pi’s IP address (again, discovered using hostname -I). Or, in VLC running on the desktop, go to File > Open Network and enter the

same address: tcp/h264://my_pi_address:8000/

Stream stills

Now let’s stream camera stills taken at regular intervals in a variation on a standard

time-lapse setup. Entered on a remote computer (which could be another Raspberry Pi), the

server script, ch15listing4.py, starts a socket to listen for a connection from your Raspberry

Pi with the camera. At the top, we import the required libraries; here we’re using PIL (you

can install it using sudo pip install pillow) to read JPEG les, but alternatives include

OpenCV and GraphicsMagick. The script then checks the image length and, if it is not zero,

constructs a stream to hold the image data and then reads it from the connection. The

image.show() command will open each image in the default image viewer: it can create a

lot of windows if left going for a while! Now to create a client script…

Stills client script

On your Raspberry Pi with the camera, the client script, ch15listing5.py, sends a

continual stream of images to the server. We’ll use a very simple protocol for communication:

rst, the length of the image will be sent as a 32-bit integer (in little-endian format), then this

will be followed by the bytes of image data. If the length is 0, this indicates that the connection

should be closed as no more images will be forthcoming. As before, for connecting the socket,

you should replace my_server in the script with the host name of the remote computer. We then

make a le-like object out of the connection. Before constructing the stream, we start a preview

to let the camera warm up for two seconds. Further down, the line if time.time() - start

> 30: limits the streaming period to 30

seconds, though you can alter this.

Note that the server script should be run

rst to ensure there’s a listening socket

ready to accept a connection from the

client script.

Taking it further, you may want to add a

way of closing each image window before

the next is generated. Also, rather than

simply showing the images, you could use

the numerous functions of PIL to process

them (see effbot.org/imagingbook).

Multiple stills are streamed to the remote

computer and displayed

95Chapter 15 Live-stream video and stills

Page 96

ch15listing1.py / Python 3

import socket import subprocess

# Start a socket listening for connections on 0.0.0.0:8000 # (0.0.0.0 means all interfaces)

server_socket = socket.socket() server_socket.bind(('0.0.0.0', 8000)) server_socket.listen(0)

# Accept a single connection and make a file-like object out of it

connection = server_socket.accept()[0].makefile('rb')

try:

# Run a viewer with an appropriate command line. Uncomment

the mplayer # version if you would prefer to use mplayer instead of VLC

cmdline = ['vlc', '--demux', 'h264', '-'] #cmdline = ['mplayer', '-fps', '25', '-cache', '1024', '-'] player = subprocess.Popen(cmdline, stdin=subprocess.PIPE) while True: # Repeatedly read 1k of data from the connection

# and write it to the media player's stdin

data = connection.read(1024) if not data: break player.stdin.write(data)

finally:

connection.close() server_socket.close() player.terminate()

ch15listing2.py / Python 3

import socket import time import picamera

# Connect a client socket to my_server:8000 # (change my_server to the hostname of your server)

client_socket = socket.socket() client_socket.connect(('my_server', 8000))

96 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

Page 97

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

ch15listing2.py (cont.) / Python 3

# Make a file-like object out of the connection

connection = client_socket.makefile('wb')

try:

camera = picamera.PiCamera() camera.resolution = (640, 480) camera.framerate = 24 # Start a preview and let the camera warm up camera.start_preview() time.sleep(2) # Start recording, sending the output to the connection for 60

# seconds, then stop

camera.start_recording(connection, format='h264') camera.wait_recording(60) camera.stop_recording()

finally:

connection.close() client_socket.close()

DOWNLOAD

magpi.cc/cameragit15

ch15listing3.py / Python 3

import socket import time import picamera

camera = picamera.PiCamera() camera.resolution = (640, 480) camera.framerate = 24

server_socket = socket.socket() server_socket.bind(('0.0.0.0', 8000)) server_socket.listen(0)

# Accept a single connection and make a file-like # object out of it

connection = server_socket.accept()[0].makefile('wb')

try:

97Chapter 15 Live-stream video and stills

Page 98

ch15listing3.py (cont.) / Python 3

camera.start_recording(connection, format='h264') camera.wait_recording(60) camera.stop_recording()

finally:

connection.close() server_socket.close()

ch15listing4.py / Python 3

import io import socket import struct from time import sleep from PIL import Image

# Start a socket listening for connections on 0.0.0.0:8000 # (0.0.0.0 means all interfaces)

server_socket = socket.socket() server_socket.bind(('0.0.0.0', 8000)) server_socket.listen(0)

# Accept a single connection and make a file-like object out of it

connection = server_socket.accept()[0].makefile('rb')

try:

while True: # Read the length of the image as a 32-bit unsigned int.

# If the length is zero, quit the loop

calcsize('<L'

if not image_len: break # Construct a stream to hold the image data and read the image

# data from the connection

image_stream = io.BytesIO() image_stream.write(connection.read(image_len)) # Rewind the stream, open it as an image with PIL

# and show it in the default image viewer

image_stream.seek(0)

98 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

image_len = struct.

)))[0]

unpack('<L'

, connection.

read

(struct.

Page 99

THE OFFICIAL RASPBERRY PI CAMERA GUIDE

ch15listing4.py (cont.) / Python 3

image = Image.open(image_stream) image.show()

finally:

connection.close() server_socket.close()

ch15listing5.py / Python 3

import io import socket import struct import time import picamera

# Connect a client socket to my_server:8000 # (change my_server to the hostname of your server)

client_socket = socket.socket() client_socket.connect(('my_server', 8000))

# Make a file-like object out of the connection

connection = client_socket.makefile('wb')

try:

camera = picamera.PiCamera() camera.resolution = (640, 480)

# Start a preview and let the camera warm up for

camera.start_preview() time.sleep(2)

# Note the start time and construct a stream to

# hold image data temporarily (we could write it # directly to connection but in this case we want # to find out the size of each capture first to keep # our protocol simple)

start = time.time() stream = io.BytesIO()

99Chapter 15 Live-stream video and stills

Page 100

ch15listing5.py (cont.) / Python 3

for foo in camera.capture_continuous(stream, 'jpeg'):

# Write the length of the capture to the stream

# and flush to ensure it actually gets sent

connection.write(struct.pack('<L', stream.tell())) connection.flush()

# Rewind the stream and send the image data over the wire

stream.seek(0) connection.write(stream.read())

# If we've been capturing for more than 30 seconds, quit

if time.time() - start > 30: break

# Reset the stream for the next capture

stream.seek(0) stream.truncate()

# Write a length of zero to the stream to signal we're done

connection.write(struct.pack('<L', 0))

finally:

connection.close() client_socket.close()

100 THE OFFICIAL RASPBERRY PI CAMERA GUIDE

RaspBerry Camera Guide

Specifications and Main Features

Frequently Asked Questions

User Manual