Method

Specifies the algorithm used to measure similarity between the sample image and the document image during shape detection.

Remarks

The Method enumeration determines how Shape Detection calculates the proximity (similarity) between a sample image and regions of the document image. The choice of method can significantly affect detection accuracy, robustness to noise, and performance.

• Use 'SAD' for fast, general-purpose detection on clean images.
• Use 'CrossCorr' for improved robustness to lighting and contrast variations, especially when the document background is inconsistent.
• Use 'SSD' for cases where outlier differences should be penalized more heavily, but be aware it may be more sensitive to noise.

Review diagnostic output to evaluate the effect of each method and select the one that best matches your document types and image quality.

Can be one of the following values:

Used By

Shape Detection