This product is a vision sensor for performing image processing measurement
through a controller of objects photographed using a camera.By connecting
an external device such as a PC, measurement commands can be input and
measurement results can be output from the external device.
1-1-1Basic Configuration of FZ3 Series
SECTION 1
Before Operation
*1: The touch pen is an accessory of a controller.
*2: FZ-SC2M and FZ-S2M cannot be connected with the FZ3-3 /FZ3-H3 controllers.
*3: Lenses for small-size cameras are required for small-size 300,000-pixel cameras.
*4: FZ-SC5M and FZ-S5M can only be connected with the FZ3-70 /FZ3-H70 controllers.
3
Preparing Controllers and CamerasSection 1-2
1-2Preparing Controllers and Cameras
1-2-1Preparing Controllers
Please check that the controller is switched on and that the Main screen is
displayed.
The first time the program is started up, the Language Setting window is displayed, so select the language.
1-2-2Adjusting Cameras
1. Tap [] of „Image mode“ in [Image display] of the Main screen Control area, and select „Through“.
The through images captured from the camera are viewed in the Image
Display area.
Note The same operation is available by tapping [View] - [Image mode] - [Through].
2. Adjust the position of measurement objects so that they display at the center of the monitor.
3. Adjust the focal distance of the lens.
When using an auto-focus camera or an intelligent camera, focus and the
iris can be automatically adjusted.
Note If a camera is used together with a lens, turn the focus ring of the lens to
adjust the focus.
The light intensity of an intelligent camera can be adjusted from the controller.
Important When using a small-size digital camera, check that the model and serial num-
ber of the camera head and camera amplifier match.When a camera head
and camera amplifier of different models and serial numbers are connected,
they may not operate correctly.
4
Software OverviewSection 1-3
1-3Software Overview
Application-oriented measurement can be configured by combining processing items or changing the settings of processing items.
5
Software OverviewSection 1-3
6
2-1What is a Scene
A combination of processing items is called a „scene“ and scenes can be easily created by combining processing items that are suited to the measurement
purpose from the list of processing items provided.
SECTION 2
Basic Operation
7
What is a SceneSection 2-1
2-1-1Scene Examples
2-1-2Switching Scenes
1. Tap „Scene switch“ in the toolbar on the Main screen.
The Switch Scene window is displayed.
Note The same operation is available by tapping [Scene] menu - [Scene switch].
2. Tap [ ] to select the scene to switch.
To switch a scene group, tap [Switch], then tap [ ] in the displayed window to select the scene group to switch.
3. Tap [OK].
The scene switches.
8
Create a SceneSection 2-2
2-2Create a Scene
In the Edit Flow window, editing buttons in the window can be used to change
the order of processing units within the scene or to delete processing units.
Specifying the position for a processing unit
and adding it ( ) ( )
Moving a processing unit ( ) ( )
Copying and pasting a processing unit ( ) ( )
Deleting a processing unit ( )
Changing the name of a processing unit ( )
Setting details of a processing unit ( )
9
Create a SceneSection 2-2
2-2-1Property window of Processing items
This window is used for detailed setting of measurement parameters and
judgement conditions for processing items. All of Processing items have similar layout.
a) Item Tab Area
Displays the settings items for the processing unit currently being set.
Perform settings starting with the item on the left.
b) Detail Area
Set detailed items.
c) Image Display Area
Displays camera images, figures, and coordinates.
d) Zoom Browser Area
Zooms in and out from the displayed image.
10
Processing Item GuidelineSection 2-3
2-3Processing Item Guideline
Define and perform measurement with display of the results.
1. Tap [Edit flow] in Toolbar.
The Edit Flow window is displayed.
2. Select a processing item to be added from the processing item tree.
3. Tap [Append].
4. To continue to add processing units. Repeat step 3.
11
Processing Item GuidelineSection 2-3
5. Either tap the icon of the processing unit to be set or tap the Set button.
The property setting window is displayed.Set detailed conditions. The displayed contents vary depending on the processing item.
6. Set conditions.
The displayed contents vary depending on the processing item.
12
Processing Item GuidelineSection 2-3
2-3-1Position Compensation
13
Processing Item GuidelineSection 2-3
2-3-2Locating (Measurement Objects Not Inclined)
14
Processing Item GuidelineSection 2-3
2-3-3Locating (Measurement Objects Inclined)
2-3-4Internal and External Inspection
15
Processing Item GuidelineSection 2-3
2-3-5Presence Inspection
2-3-6Dimension Inspection/Measurement
16
Processing Item GuidelineSection 2-3
2-3-7Burr Inspection
2-3-8Text Comparison/Inspection
17
Processing Item GuidelineSection 2-3
2-3-9Defect/Contamination Inspection
2-3-10 Quantity Inspection/Measurement
18
Processing Item GuidelineSection 2-3
2-3-11 Inspection for Presence of Different Objects
2-3-12 Hole Position Measurement
2-3-13 Selecting Measurement Processing Items According to the
Measurement Method and Purpose
This section describes methods for selecting processing items appropriate to
different measurement objectives such as counting quantities, checking for
deformation, and checking for contamination.
19
Processing Item GuidelineSection 2-3
20
3-1Search
SECTION 3
Processing Item Overview
Register the feature sections of the measurement object as an image pattern
(model), then find the most similar part to these models from the input images
to detect the position.
The correlation value showing the degree of similarity, measurement object
position, and inclination can be output.
• When identifying the shape of measurement objects (for detecting defects
or foreign matter)
21
Flexible SearchSection 3-2
3-2Flexible Search
In Flexible Search, multiple measurement object features (models) are registered beforehand.Parts from input images that most resemble the multiple
models are searched for, and correlation (similarity) and position are determined.
• To treat models with only slight variations as the same and prevent excessive filtering out.
3-2-1Model Registration (Flexible Search)
Register the parts to measure as the model.
A total of 5 models, 0 through 4, can be registered, with no restriction on the
size.
If a model has different printing qualities and shapes, more than one models
should be registered.
The position at the time of registration is also registered in the model informa-
tion.Place the measurement object in the correct position when registering a
model.
22
Sensitive SearchSection 3-3
3-3Sensitive Search
The registered models are automatically finely divided and matched in
detail.Of the divided models, the one with the lowest correlation is output.
Sensitive search is suitable when the difference between the model image
and measurement image is small and regular searches do not produce differences in correlation.
• When identifying the shape of the divided area
Setting itemSet value [factory default]Description
Sub-model number X 0 to 10 [4]This sets the number of divi-
sions of the registered model in
the X direction.
Sub-model number Y 0 to 10 [4]This sets the number of divi-
Stab.1 to 15 [The default value
depend on the connected
camera.12 or 15]
Prec.1 to 3 [2]Specify which is to have priority,
Plain inspection• Checked
• [Unchecked]
sions of the registered model in
the Y direction.
Specify which is to have priority,
measurement stability or speed.
If lowering stability does not
speed up processing, it is likely
that many candidates have
been detected.In this case,
specify a larger value for
„Candidate LV“ or „Stab.“
measurement positional precision or speed.
Specify whether or not to
inspect the plain region.
23
ECM SearchSection 3-4
3-4ECM Search
This processing item searches the input image for parts having a high degree
of similarity to the target mark (model), and measures its correlation value
(similarity) and position.
In a normal search, image pattern models are used that look at the color and
light/dark information, but in an ECM search, models are used that look at the
profile information.Therefore, this processing assures a reliable search even
for low-contrast or noisy images.
• To measure the location of a mark
24
EC Circle SearchSection 3-5
3-5EC Circle Search
This processing item searches the input image for parts having a high degree
of similarity to the target circle mark (model), and measures its circle evaluated value (similarity) and position.In a normal search, image pattern models
are used that look at the color and light/dark information. In EC Circle Search,
however, models are used that look at the profile.Therefore, this processing
assures a reliable search even for low-contrast or noisy images.It is also possible to measure the number of circles in the input image.
• This counts how many circles there are of the specified size.Since circles
are extracted using the shape information in „Round“, the circles being
deformed or dirty does not affect counting.
3-5-1Measurement Parameters
Setting itemSet value [factory default]Description
Search type[Single search]This is set when there is one
Setting itemSet value [factory default] Description
Radius1 to 9999This item sets the radius of the
Radius range[1] to 9999This measures the measured
search target.
Multi searchThis is set when there is more
than one search target.
circle measured.
This is displayed on the screen
with a solid blue line.
circle radius ± the permitted
radius width.
This is displayed on the screen
with a broken blue line.
25
ClassificationSection 3-6
3-6Classification
• When various kinds of products on a production line need to be classified
and identified
3-6-1Model Registration (Classification)
Pre-register as models the sections to be used as reference for classification.
Models can be registered with any of 36 indexes, from 0 to 35, and up to 5
models can be registered for each index.
When there is variation among the model print quality and shapes, pre-regis-
ter multiple models for the same index.
26
Edge PositionSection 3-7
3-7Edge Position
This processing item detects the position of the measurement object by using
the change in color within the measurement region.
• To calculate edge coordinates of measurement objects
• To find the width of a measurement object
Using a Expression, the width of a measurement object can be calculated
from the difference between two edge positions.
• The edge is scanned from the start point of the area toward the end point.
When setting up the measurement region, pay attention to the detection di-
rection of the edge.
• Use the zoom function if the measurement region is too small to identify
the direction of the arrow.
27
Edge PitchSection 3-8
3-8Edge Pitch
Finds and counts the edges by measuring the color change within the measurement region.
• When calculating number of pins of IC or connectors
• To calculate the pin width and the distance (pitch) between mid-points
between two pins
• When setting up a measurement region, please include all the edges to
be detected.
28
Scan Edge PositionSection 3-9
3-9Scan Edge Position
This processing item detects the position of the measurement object by using
the change in color within the measurement region.By dividing the measurement region, the following effects can be expected compared to ordinary edge
position measurement.
• Detailed information, such as the closest point or furthest point from the
measurement start point, can be calculated.
• The inclination or degree of unevenness of the measured object can be
calculated.
• To calculate multiple edge positions of the measurement object from statist
29
Scan Edge PositionSection 3-9
3-9-1Judgment Conditions (Scan Edge Position)
30
Scan Edge WidthSection 3-10
3-10Scan Edge Width
This processing item detects the position of the measurement object by using
the change in color within the measurement region.By dividing the measurement region, you can get the following values.
• Local width of the work
• Average width of the work
• When getting several widths of a measurement object
• To find the width of a measurement object
Using a Expression, the width of a measurement object can be calculated
from the difference between two edge positions.
The region is divided equally.
31
Color DataSection 3-11
3-11Color Data
Inspect by finding the average color of the measurement region and using its
difference from the registered reference color and the color variation in the
measurement area.Alternatively, you can only detect the color tone while
neglect the effect of image brightness.
For monochrome cameras, examination is performed by measuring the difference between the average density of the measurement region and the registered reference density (density average), and the density deviation in the
measurement region (density deviation).
Used in the Following Case
• When measuring the presence of measurement objects
Setting itemSet valueDescription
Color difference0 to 442Specify the upper and lower
Color deviation0 to 221Specify the upper and lower
3-11-1 For Monochrome Cameras:
Setting itemSet valueDescription
Density average0 to 255Specify the upper and lower
Destiny deviation0 to 127Specify the upper and lower
32
limit values for the difference
between the average color of
the measurement region and
the reference color.
limit values for the deviation of
the average color in the measurement region.
limit values for judging the average density of the measurement region.
limit values for the deviation of
the average density in the measurement region.
Gravity and AreaSection 3-12
3-12Gravity and Area
Inspect using the area of the specified color.
Used in the Following Case
• Label deviation measurement
• Detection of defects, contamination, and stains of measurement objects
whose appearance is not defined
Setting itemSet valueDescription
Area0 to 999999999.9999Specify the area to be judged
as OK.
Gravity X-99999.9999 to 99999.9999 Specify the range of X-axis
shifting that is judged to be OK.
Gravity Y-99999.9999 to 99999.9999 Specify the range of Y-axis shift-
ing that is judged to be OK.
33
LabelingSection 3-13
3-13Labeling
You can count the number of labels with a specified color or find the area and
center of gravity of a specified label number.
Used in the Following Case
• Label count inspection
34
LabelingSection 3-13
Setting itemSet value [Factory default]Description
Filling up holes• Checked
• [Unchecked]
Select the process method for
the part encircled by the designated color circle.
When checked, the hole is processed as having the specified
color.
Outside trimming• Checked
• [Unchecked]
Object area range0 to 999999999Specify the range of the area to
Sort condition • Area ascending
• [Area descending]
• X ascending
• X descending
• Y ascending
• Y descending
Label No.[0] to 2499Input the label number for the
This option can be used only
when there is a section of the
designated color in the measurement region that does not
need to be measured.
When „Checked“ is set, the
whole area outside of the measurement region is extracted as
having the specified color.
be judged as a label.
Specify the conditions by which
label number is re-assigned.
When sorting referencing the X
and Y coordinates, the upper
left is the origin.
data to be output.
35
Label DataSection 3-14
3-14Label Data
You can specify a desired label number and obtain measurement values for
that label stored by other processing units.
The processing items that can be set up as reference objects are the following
items that perform the labeling processing.
• Labeling
Used in the Following Case
• Label position acquisition
36
Note Do not insert the following processing items between Label Data and Labeling
units.
• Camera Image Input
• Camera Switching
• Position Compensation
• Color Gray Filter
• Filtering
DefectSection 3-15
3-15Defect
Detect defects and contamination using color variation within the measurement region.
This is real color processing, so even if defect and contamination colors
change or the background color changes, stable inspection is possible.
Used in the Following Case
• Detecting defects, contaminations and spots on plain measurement
objects
• Measure appearance defects and defects of parts
37
DefectSection 3-15
3-15-1 Region Setting (Defect)
PTDescription
Wide lineSelected when detecting defects and burrs of the measure-
ment objects.
Wide circle, wide arc Selected when detecting defects and burrs of the circle
measurement objects.
Rectangle, ellipse
(circle), polygon
Selected when detecting the overall defects of specified zones
and measurement objects.
Defect detection mechanism
• After measurement region is drawn, a rectangle (defect detection region)
is automatically formed in this region.While moving the defect detection
region around, calculate the RGB color averages at each location and find
the defect detection difference with surrounding defects. This difference is
called the defect level. Calculate the defect level for all defect detection
areas. If the maximum value exceeds the judgement value, it is judged
that there are defects in the measurement region.
38
DefectSection 3-15
ItemSet value [Factory default]Description
Defect judgement0 to 999 [100] Specify the upper limit for defect
judgement. (The lower limit is
fixed at 0.)
When „30“ is set, the OK value
should be within the range of 0
to 30.
Area judgement0 to A_MAX Specify the maximum defect
area.
A_MAX: 307,200 for a 0.3-
megapixel camera, 1,920,000
for a 2-megapixel camera
39
Precise DefectSection 3-16
3-16Precise Defect
Defects and contamination on plain measurement objects can be detected
with high precision by performing differential processing on the image.By
changing the size of elements used for detection, comparison intervals, etc.,
fine customization of speed and precision is possible.
Setting itemSet value [Factory default]Description
Size X4 to 64 [16] Specify the X-axis size of
defects/contamination to be
detected.The higher this value,
the higher the degree of defects
for large defects. Specify in
units of pixels.
Size Y4 to 64 [16] Specify the Y-axis size of
Sampling interval X1 to 64 [2]Specify the interval for creating
Sampling interval Y1 to 64 [2] Specify the interval for creating
Comparing interval X 1 to 32 [10] Set the number of neighboring
defects/contamination to be
detected.The higher this value,
the higher the degree of defects
for large defects.Specify in units
of pixels.
elements along the X axis.The
smaller this value, the greater
the defect detection performance, but the slower the processing speed.Specify in units
of pixels.
elements along the Y axis. The
smaller this value, the greater
the defect detection performance, but the slower the processing speed.Specify in units
of pixels.
elements compared with when
the degree of defect is calculated,For example, if the Sampling interval X is set to 4 and
the comparing interval X is set
to 2, comparison is with separate elements of 4 x 2 = 8 pixels
along the X axis.
40
Precise DefectSection 3-16
Setting itemSet value [Factory default]Description
Comparing interval Y 1 to 32 [10] Set the number of neighboring
elements compared with when
the degree of defect is calculated, For example, if the sampling interval Y is set to 4 and
the comparing interval Y is set
to 2, comparison is with separate elements of 4 x 2 = 8 pixels
along the Y axis.
Direction • X (circumferential)
•Y (radial)
• Diagonal
Set the direction for detecting
defects.
The smaller the direction setting
count, the shorter the processing time.
41
Fine MatchingSection 3-17
3-17Fine Matching
Differences can be detected in a fast and highly precise way by overlapping
registered fine images with input images (matching).
Used in the Following Case
• To precisely detect trivial defects at the edges of text and patterns
42
Circle AngleSection 3-18
3-18Circle Angle
• To correct the tilting of circle measurement objects
• When drawing the measurement region, the featured part should lie on
the circumference.
43
Model DictionarySection 3-19
3-19Model Dictionary
Register a model to use for [Character Inspection].
Model data registered in the [Model Dictionary] can be referred to from multi-
ple [Character Inspection] items in the same scene.
• To create the dictionary to be used for Character Inspection and Date
Verification
3-19-1 Model Automatic Registration (Model Dictionary)
This method encloses a character string, cuts out one character at a time from
it and registers them as models.
1. In the Item Tab area, tap [Auto registration].
2. In the Detail area, select „Auto extract region“.
3. Specify the range to register as the auto extract region using the drawing
tools.
4. Tap [Extract model].
5. In the Detail area, select „Auto model region“.
6. Tap the model region in the Image Display area.
In the context menu, an index list is displayed.
44
7. Select the index to register to.
8. Tap [Register model].
Character InspectionSection 3-20
3-20Character Inspection
Using model images registered in a [Model Dictionary], this processing item
performs character recognition by correlation searches.
• When identifying standard character data (check of product model name)
Setting itemSet value [Factory default]Description
Inspection mode • [OCR]
• OCR + Count
•OCV
Direction
[
]
Select the inspection mode of
characters.
• OCR: The character string is
read in.
• OCR + Count: The character
string is read in.Also, the
number of characters is
inspected.
• OCV: Inspects whether or not
the same characters are lined
up as the preset correct character string combination.
Specify the direction of character reading.
Horizontal
succession
Vertical succession[0] to 99 Specify the allowable overlap-
Dictionary candidate • [Unchecked] (Not used)
Rough candidate0 to 100 [40] When „Dictionary candidate“ is
Detail candidate0 to 100 [60] When „Dictionary candidate“ is
[0] to 99 If the characters are too close
together to read in well,
increase this.Specify the allowable overlapping range to be
read for candidate points. This
item is enabled when „Direction“
is „“ or „“.
ping range to be read for candidate points.This item is enabled
“ or „“.
• Checked (Used)
when „Direction“ is „
Specify whether to use candi-
date point levels specified in the
Model Dictionary or not.
unchecked, specify a value for
the Rough candidate.
unchecked, specify a value for
the Detailed candidate.
45
Date VerificationSection 3-21
3-21Date Verification
This processing item creates a target string from the current date/time and
compares it with read-in strings.
• When inspecting date of manufacture
46
LabelDescription
0 to 9Normal numeric value input
A to ZNormal alphabet input
‘ - . : /Normal mark input
*Character presence judgement
$Number judgement
mYYThe last two digits of the current year
mYYYYFour digits of the current year
mHHTwo digits of the current year in the Japanese Heisei calendar
mMMCurrent month
mDDCurrent day
Date VerificationSection 3-21
LabelDescription
mRRCurrent hour
mNNCurrent minute
vYYThe last two digits of the year after a set period of time
vYYYYFour digits of the year after a set period of time
vHHTwo digits of the year after a set period of time in the Japanese
Heisei calendar
vMMMonth after a set period of time
vDDDay after a set period of time
eY1Encrypted year 1
eM1Encrypted month 1
eD1Encrypted day 1
eR1Encrypted hour 1
eN1Encrypted minute 1
eY2Encrypted year 2
eM2Encrypted month 2
eD2Encrypted day 2
eR2Encrypted hour 2
eN2Encrypted minute 2
47
Date VerificationSection 3-21
48
4-1Image Logging
This is used when saving measurement images to on-board memory, RAMDisk or USB memory.
This enables preparation of logging conditions using an expression and is
more flexible than the system image logging conditions settings.
However, the settings of this unit are enabled if „None“ is set on the [Logging
setting] of the main screen [Measure] menu.
If settings that perform image logging for multiple units during measurement
are executed, the last settings executed are enabled.
SECTION 4
Support measurement functions
• This is used when saving logging images under specific conditions.
4-1-1Logging Conditions (Image Logging)
Setting itemSet value [Factory default]Description
Logging condition • [None]No images are saved.
• Only NG Saves images only if an NG
• AllAll measured images are saved.
occurs.If an NG occurs downstream from the image logging
processing unit, image logging
is not performed.
Insert image logging as close to
the end of the scene as possible
49
Image LoggingSection 4-1
50
5-1Data Output
SECTION 5
Output result
• Output data to the external devices such as programmable controller and
PC with the no-order mode via the serial interface.
5-1-1Settings (Data Output)
Set up the output contents with the expression.
Up to 8 expressions including 0 to 7 can be set in each unit.
Set value
[factory default]
[RS-232C/RS-422]Communication is performed via a RS-232C/RS-422
EthernetCommunication is performed via the Ethernet.
Set value [factory
default]
[ASCII]Outputs in the ASCII format.
BinaryOutputs as binary data. Measurement values are multiplied by
Description
connection.
Description
1000 and output is continuous with 4 bytes per each data item.
51
Data OutputSection 5-1
52
6-1Position Compensation
The positional deviation of measurement objects can be corrected using measured values saved by other processing units. Compare the measured coordinates with the reference coordinates of the applicable processing unit, and
move the image by the amount of the difference.
• Even with different positions for the same measurement object, correct
measurement can still be performed by correcting the position of the input
image.There is no need to reposition the measurement object itself.
SECTION 6
Compensate image
53
Revision history
Cat. No. Q29E-EN-01
Revision code
A manual revision code appears as a suffix to the catalog number on the front cover of the manual.
The following table outlines the changes made to the manual during each revision. Page numbers refer to the
previous version.
Revision codeDateRevised content
01November 2009Original production
54
Loading...
+ hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.