Machine Vision Technology
Inspection Functions Basics
Area sensor (binary conversion)
This function detects the area by counting the number of white or black pixels within the measurement window. With color processing, the area is detected by counting the number of extracted or unextracted color pixels.
Dimension measurement
A variety of dimensions are measured by setting a window on the image and measuring the distance between the detected edges.
Inner diameter measurement
|
|
Outer diameter measurement
|
Blob
A group of pixels with the same intensity range within a binary (black and white) image is called a "blob". The quantity of blobs, and features such as the area and center point, are measured by the blob tool.
Features Measured by Blob
Center of Gravity
The center of gravity of the blob is measured using the X-Y coordinate. |
|
Area
The number of pixels in the blob is measured. |
Principal Axis Angle
The principal axis angle from the horizontal (X) axis is measured
within the range of ±90°.
|
|
Feret Diameter
The lengths in X and Y directions of the circumscribed rectangle of the blob are measured.
|
Perimiter
The perimiter of the blob is measured.
|
|
Roundness
A perfect circle is represented by the number of 1.000 to measure against the roundness of each blob.
|
Detection of Scratches and Flaws (STAIN)
Detection Principle of Stain (when detecting direction is set to X)
- By moving a segment of arbitrary size within the inspection area by 1/4 of the size (amount of movement), the mean intensity is measured.
- The difference between the maximum and minimum intensity within the 4 directional segments including the observation segment (
 in the figure to the right) is measured. This value becomes the "stain level" of the observation segment.
- If the stain level exceeds the preset threshold, the observation segment is counted as a stain. This count will be reflected in the inspection result, called the "stain area".
Then, the observation segment is moved by 1/4 of its size again and the procedure is repeated from 1 to 3. |
|
|
Reference
When detection directions are specified as X and Y (two-dimensional) directions
If X and Y are specified, the difference between the maximum and minimum intensity within the 16 segments (4 segments in X direction x 4 segments in Y direction including the observation segment) is measured. |
|
|
Advantages of Stain Algorithm in Comparison with a Standard Method Using Binarization
- Because a grayscale picture is directly processed, there is no need to configure
binary settings according to the target.
- Since the change in contrast between the observation segment and the background is used instead of the absolute intensity values, the measurement remains robust even as surface conditions and illumination intensity change.
Detection of scratches on the surface of cast part
 |
|
 |
Initial condition |
|
Stain Tool: The scratch can be detected even when illumination intensity is lowered. |
| |
|
 |
| |
|
Blob Tool: The binarized image changed due to lowered illumination intensity and a false detection is made. |
Trend Edge Position
The specified segment in the edge detection area is scanned and maximum, minimum and average values of each point are output. This is effective for the measurement of maximum/minimum external diameters and detection of protruding parts.
When the measurement area is in rectangular shape
|
|
When measurement area is in circular shape
|
- If highly accurate position detection is required, reduce the segment size.
- In order to shorten the processing time, increase the shift width of the segment (amount of movement).
- The trend direction signifies the direction towards which the segment is moved.
Intensity Inspection
Because contrast (intensity) mean value, maximum/minimum values, and deviation value within the measurement area can be measured, it can also be used for illumination intensity fluctuation-resistant area measurement or high-speed scratch detection.
When temperature difference is large
- Mean intensity of window 1: 50
- Mean intensity of window 2: 200
- Intensity difference: 150
|
|
When temperature difference is small
- Mean intensity of window 1: 150
- Mean intensity of window 2: 200
- Intensity difference: 50
|
New Topics and Guidebooks
German
French
English
