Pixel Difference Detection

## Pixel Difference Detection

Used to detect image defects. First generate a template image, calculate the gray difference between the image to be detected and the template image, and extract the defect area by gray threshold extraction on the difference image

## Module Principle
Two images of the same size perform gray difference set, and region extraction is performed on the difference image

## Usage Scenarios
Detect defects

## Usage Method
1. Learn template first
![](/media/202602/2026-02-06_135149_2284450.7206526586851469.png)

2. Set mask

![](/media/202602/2026-02-06_135338_1850920.6549451274317847.png)

3. Set detection parameters

![](/media/202602/2026-02-06_135406_9227950.0318277726750108.png)

4. Result display, the center of the object to be detected does not have a white band, which is darker than the template, so it is detected.
![](/media/202602/2026-02-06_135507_0657170.03613051813816415.png)
## Parameter Configuration Description
Gray Threshold: When the absolute difference between the image to be detected and the template image is within the set threshold interval, the region is detected.
Defect Type: Brighter than template, Darker than template, All;
Erosion, Dilation: Perform erosion and dilation operations on the detected regions.
Screening: Perform feature value screening on the detected regions to screen out regions that meet the conditions, see Blob Detection.
Learn Template: Click to learn the image in the ROI as the template image.
![](/media/202602/2026-02-06_135748_3184160.712266701755668.png)
## Output Description
Running Result: Output True if there is a difference, output False if there is no difference.
Rotated Rectangle ROI: Output of ROI used by the module.
![](/media/202602/2026-02-06_140314_8521830.050996623350128556.png)

---

Click here to jump to the corresponding Chinese manual page: [Pixel Difference Detection](/doc/74)