BinarizationMethod

The 'Simple' method applies a single, user-defined threshold to all pixels in the image.

• Pixels above the threshold become white; those below become black.
• Best for images with consistent lighting and strong contrast.
• Fastest method, but may not handle shadows or gradients well.

Example:
Use for clean, high-contrast forms or barcodes.

The 'Auto' method analyzes the image histogram to find the optimal global threshold using Otsu's algorithm.

• Automatically adapts to most document types and lighting conditions.
• Additional properties allow you to adjust or constrain the detected threshold.
• Recommended for most scanned documents and general-purpose binarization.

Example:
Use for mixed batches of documents where lighting and background may vary.

The 'Adaptive' method calculates a local threshold for each pixel using a sliding window.

• Handles shadows, gradients, and uneven backgrounds by adapting to local image content.
• Window size and kernel type can be configured for best results.
• May increase processing time compared to 'Simple' or 'Auto'.

Example:
Use for receipts, photographs, or documents with variable lighting.

The 'Dynamic' method is similar to 'Adaptive', but more aggressively preserves dark regions.

• Useful for retaining faint handwriting, stamps, or marginalia that might be lost with other methods.
• Includes a 'Black Grouping' property to control how much dark content is preserved.
• May introduce more background noise if set too aggressively.

Example:
Use for archival documents, forms with pencil notes, or pages with stamps.