FLIR LWIR Application Note
FLIR Camera Adjustments
FLIR Commercial Systems
70 Castilian Drive
Goleta, CA 93117
Phone: +1.805.964.9797
www.flir.com
LWIR Video Camera
Application Note
Document Number: 102-PS242-100-01
Version: 110
Issue Date: June 2014
FLIR Camera Adjustments
Table of Contents
FLIR Camera Adjustments ........................................................................................................................... 1
LWIR Video Camera .................................................................................................................................... 1
Application Note ........................................................................................................................................... 1
Document Number: 102-PS242-100-01 ....................................................................................................... 1
1.0 Document .......................................................................................................................................... 4
1.1 Revision History ........................................................................................................................... 4
1.2 Scope ............................................................................................................................................. 4
2.0 Automatic AGC Parameters.............................................................................................................. 5
2.1 Introduction to Histograms ........................................................................................................... 6
2.2 Linear Histogram .......................................................................................................................... 7
2.3 Plateau Histogram Equalization .................................................................................................... 8
2.4 Information-based and Information-based equalization ............................................................. 20
2.5 Legacy AGC modes .................................................................................................................... 21
2.5 Digital Data Enhancement (DDE) .............................................................................................. 22
3.0 LUT Palettes and Polarity ............................................................................................................... 24
4.0 FFC Warning Indicator ................................................................................................................... 27
Table of Figures
Figure 1: 14-bit Histogram ............................................................................................................................ 6
Figure 2: Image of scene ............................................................................................................................... 7
Figure 3: Linear AGC, ITT Mean: 127, Max Gain: 8 ................................................................................... 7
Figure 4: Illustration of the Linear-Histogram Mapping Function ............................................................. 8
Figure 5: Plateau: 150, ITT Mean: 127, Max Gain: 8 ................................................................................... 9
Figure 6: Image Transform Table for Linear and Plateau algorithms......................................................... 10
Figure 7: Plateau: 250, ITT: 110, Max Gain: 8 ........................................................................................... 11
Figure 8: Plateau: 250, ITT: 150, Max Gain: 8 ........................................................................................... 11
Figure 9: Low contrast scene in 14-bit space. ............................................................................................. 12
Figure 10: Plateau: 250, ITT: 127, Max Gain: 8 ......................................................................................... 12
Figure 11: Low contrast Scene: default settings ......................................................................................... 12
Figure 12: Plateau: 250, ITT: 127, Max Gain: 25 ....................................................................................... 13
Figure 13: Low contrast scene: high gain ................................................................................................... 13
Figure 14: Plateau: 250, ITT: 127, Max Gain: 50 ....................................................................................... 13
Figure 15: Low Contrast Scene: very high gain ......................................................................................... 13
Figure 16: Illustration of Plateau Value .................................................................................................... 14
Figure 17: Illustration of Maximum Gain in a Bland Image .................................................................... 15
Figure 18: Illustration of ITT Midpoint .................................................................................................... 16
FLIR Camera Adjustments
102-PS242-100-01 Rev110
June 2014
Page 3 of 28
Figure 19: Illustration of Active Contrast Enhancement (ACE) ................................................................. 17
Figure 20: Illustration of Smart Scene Optimization (SSO) ....................................................................... 18
Figure 21: Illustration of ROI ................................................................................................................... 19
Figure 22: Illustration of the difference between Plateau Equalization, Information-based, and
Information-based Equalization algorithms ................................................................................................ 21
Figure 23: Illustration of Information Threshold ........................................................................................ 21
Figure 24: Illustration of Noise Suppression with DDE ........................................................................... 23
Figure 25: Illustration of Detail Enhancement with DDE ........................................................................ 23
Figure 26: White Hot .................................................................................................................................. 24
Figure 27: Black Hot ................................................................................................................................... 24
Figure 28: Look-Up Table Options (Without Isotherms) ........................................................................... 25
Figure 20: Isotherm LUT Scale Example ................................................................................................... 26
Figure 21: Look-Up Table Options (with Isotherms) ................................................................................. 27
Figure 29: FFC Warning ............................................................................................................................. 28
FLIR Camera Adjustments
102-PS242-100-01 Rev110
June 2014
Page 4 of 28
Version
Date
Comments
100
10/25/2011
Initial Release
110
6/20/2014
Updates for the Tau 2.7 and Quark 2 release
Document Title
Document
Number
Description
Tau Quick Start Guide
102-PS242-01
Quick Start Guide for first-time use
Quark Quick Start Guide
102-PS241-01
Quick Start Guide for first-time use
FLIR Camera Controller GUI
User’s Guide
102-PS242-02
Detailed Descriptions for functions and adjustments
for FLIR cameras using the FLIR Camera Controller
GUI
Tau 2 Product Specification
102-PS242-40
Product specification and feature description
Quark Product Specification
102-PS241-40
Product specification and feature description
Tau 2 Electrical IDD
102-PS242-41
Written for Electrical Engineers to have all necessary
information to interface to a Tau 2 camera
Quark Electrical IDD
102-PS241-41
Written for Electrical Engineers to have all necessary
information to interface to a Tau 2 camera
Tau 2/Quark Software IDD
102-PS242-43
Written for Software Engineers to have all necessary
information for serial control of Tau 2 and Quark
Assorted Mechanical
Drawings and Models
Various
There are drawings and 3D models for various
camera configurations for mechanical integration
Application Notes
Various
Written for Systems Engineers and general users of
advanced features such as Gain Calibration,
Supplemental FFC Calibration, NVFFC Calibration,
Camera Link, On-Screen Symbology, AGC/DDE
explanation, Camera Mounting, Spectral Response,
Optical Interface for lens design, and others.
1.0 Document
1.1 Revision History
1.2 Scope
This note is intended to provide a better understanding of FLIR image processing algorithms. Once these
are well understood by the user, the camera can be optimized to give the best possible image for a given
scenario. This document applies to the FLIR Quark, Quark 2, Tau, Tau 2 and Neutrino cameras. These
cores can be found in most FLIR Commercial Systems products.
The FLIR website will have the newest version of this document as well as offer access to many other
supplemental resources: http://www.flir.com/cvs/cores/resources/
Here is a sample of some of the resources that can be found:
There is also a large amount of information in the Frequently Asked Questions (FAQ) section on the
FLIR website: http://www.flir.com/cvs/cores/knowledgebase/. Additionally, a FLIR Applications
Engineer can be contacted at 888.747.FLIR (888.747.3547).
FLIR Camera Adjustments
102-PS242-100-01 Rev110
June 2014
Page 5 of 28
2.0 Automatic AGC Parameters
The first thing to understand is that the detector data is directly streamed from the sensor as 14-bit values
for each pixel in the array. The analog image is displayed using 8-bit values and almost all commercial
displays are 8-bit devices. In other words the video is displayed on a 0-255 scale rather than the full 016384 resolution of the sensor. This means that there must be some compression to get the data into a
format that can be displayed. Throughout this note, there are histograms that are represented in Signal vs.
Number of Pixels. A histogram is a sorting of pixel values into intensity “bins”. What this means is the
bit value (which increases as pixels get brighter) is on the x-axis and the number of pixels in the image
that have that bit value is on the y-axis. This is a way of plotting image data in order to illustrate which
are the most significant intensity values. The algorithms attempt to compress the data in a meaningful
way that preserves as much of the image content as possible.
The Tau 2 core provides multiple AGC algorithms used to transform 14-bit data to 8-bit. These options
include the following, with associated parameters shown below each algorithm:
Plateau equalization
o Plateau value
o Maximum gain
o ITT midpoint
o ACE threshold
o SSO value
o Tail rejection
o Region of Interest (ROI)
o IIR filter
Information-based and Information-based equalization
o Information-based Threshold
Linear histogram
o ITT midpoint
o ROI
o IIR filter
Manual
o Brightness
o Contrast
o IIR filter
FLIR Camera Adjustments
102-PS242-100-01 Rev110
June 2014
Page 6 of 28
1 2 3
Auto-bright
o Brightness
o Contrast
o IIR filter
Once-bright
o Brightness bias
o Contrast
o IIR filter
Note: FLIR highly recommends that each customer optimize AGC settings for each particular
application. “Preferred” AGC settings are highly subjective and vary considerably depending upon
scene content and user preferences. Generally speaking, FLIR recommends the plateau equalization
algorithm, but there are scenarios where each of the other algorithms may be better suited. The FLIR
GUI provides auto presets that can be used to tune AGC to the specific scene.
2.1 Introduction to Histograms
The following histogram is the 14-bit data taken from a Tau 320 with a cold water bottle, a midtemperature wall, and a hot coffee mug in the scene. These three objects can be seen in the data histogram
as three separate peaks. The lowest bit values, which are farthest to the left in the histogram, are the
coldest pixels in the scene. The values in 14-bit space range from 0 to 16384.
The following image is associated with this histogram. You can see the cold, black water bottle, the grey
background, and the hot, white coffee mug. You can also see that the water bottle is fairly uniform and
Figure 1: 14-bit Histogram
FLIR Camera Adjustments
102-PS242-100-01 Rev110
June 2014
Page 7 of 28
1
2
3
1 2 3
has a narrow spike whereas the mug has different temperatures in the handle and above the coffee line.
For this reason, the data is more spread in the histogram at point 3.
Figure 2: Image of scene
2.2 Linear Histogram
The first and simplest method to translate the data into 8-bit space is using a linear algorithm. Although
this algorithm is not typically used, it will help illustrate the concept of using a transfer funciton to map
from one space to another. A typical linear intensity transfer table will map the middle 90% of the
histogram to 8 bit space. The bottom and top 5% are discarded. This algorithm finds the interesting
portion of the data and crops above and below it. The following histogram is a representation of the same
scene in 8-bit space. Notice the three peaks from the 14-bit data represented in 8-bit space and that the
values on the x-axis now range from 0-255 rather than 0-16384.
Figure 3: Linear AGC, ITT Mean: 127, Max Gain: 8
FLIR Camera Adjustments
102-PS242-100-01 Rev110
June 2014
Page 8 of 28
(a) ITT Midpoint = 128
(b) ITT Midpoint = 96
14bit_5%
14bit_95%
Avg(14bit_5%, 14bit_95%)
specified
ITT Midpoint
14bit_5%
14bit_95%
Avg(14bit_5%, 14bit_95%)
specified
ITT Midpoint
The linear histogram algorithm performs a linear transformation from 14-bit to 8-bit of the form:
8biti = m * 14biti + b
The slope of the transformation is computed automatically based on the ROI histogram:
m = 255 / (14bit_(100 – Tail Rejection)% - 14bit_(Tail Rejection)%),
where 14bit_(Tail Rejection)% is the 14-bit value corresponding to the user selectable tail rejection
percentage point on the cumulative ROI histogram and 14bit_(100 – Tail Rejection)% is the value
corresponding to the difference between 100% and the user selectable tail rejection percentage point in
Tau 2.7 and Quark 2.0.
The offset is then computed as
b = ITT midpoint - avg(14bit_(100 – Tail Rejection)%, 14bit_(Tail Rejection)%),* m
In other words, the algorithm attempts to map the midway point between the top and bottom tail rejection
points on the cumulative histogram to the specified ITT midpoint, as shown in Figure 4 for the case in
which the tail rejection parameter selected is 5%. The 8-bit values resulting from the above equations are
clipped to a minimum value of 0 and a maximum value of 255.
Figure 4: Illustration of the Linear-Histogram Mapping Function
2.3 Plateau Histogram Equalization
The Plateau Histogram Equalization algorithm seeks to maximize the dynamic range available for the
content of the scene. It does this using a transfer function that is based on the number of pixels that are in
each bin and allocating more 8-bit range for that bin. The Plateau value is the pixels/bin limit when the
transfer function is maximized. When this number is small, the Automatic AGC will approach a linear
algorithm that preserves a linear mapping between the 14-bit and 8-bit data. The goal of the Automatic
algorithm is to try and make each of the 255 bins have the same number of pixels in it, which should give
the best contrast for the given scene. When the plateau value is higher, the algorithm is more able to
FLIR Camera Adjustments
102-PS242-100-01 Rev110
June 2014
Page 9 of 28
redistribute the data to achieve this goal. This prevents wasted levels of grey on regions that have no
scene content and can visually be seen in the histograms by noticing that peaks are much smoother and
the data is spread much more evenly.
Figure 5: Plateau: 150, ITT Mean: 127, Max Gain: 8
Compare to Linear Histogram in Figure 3
The following plot shows the Image Transform Table for both Linear and Plateau Histogram
Equalization. The 14-bit value on the x-axis will map to the 8-bit value on the y-axis where the
conversion is plotted. In 14-bit regions with low contrast, the curve is much flatter and there are not as
many 8-bit values consumed. In high detail regions, the curve is steep and more 8-bit values are used.
Loading...