Extract Page

The 'Binarization' property controls how the input image is converted to black and white prior to edge detection. Proper binarization is essential for reliable page extraction, as it enhances the contrast between the page and the background, making edges more distinct.

• Dark Backgrounds:
  For pages placed on a dark or black background, configure binarization to produce a solid black border and a mostly white page area. This maximizes the contrast at the page edges, improving detection accuracy.
• Light Backgrounds:
  For pages on a light or similar background, adjust binarization so that the page outline is visible as a line. The goal is to ensure that the page edges are clearly defined, even if the background is not much darker than the page itself.
• Diagnostics:
  Use the 'Binarized' diagnostic image to review the effect of your settings. If the page edges are not clearly visible, experiment with different binarization methods and parameters.
• Example:
  For a dark background, use the 'Auto' or 'Adaptive' method with default settings. For a light background, you may need to fine-tune the threshold or use a different method to enhance edge visibility.

The 'Border Size' property defines the thickness of the region along each edge of the image where the algorithm searches for page edges. Increasing the border size can help detect pages that are not centered or have wide margins, while smaller values focus detection closer to the image boundaries.

• Units:
  You can specify the border size using unit-aware values such as 0.25in, 10px, or 6mm. Each side (left, top, right, bottom) can be set independently, or you can use shorthand notation to set all sides at once.
• Diagnostics:
  The border region is annotated in the 'Zoning' diagnostic image, allowing you to visualize and adjust the detection area.
• Tuning Tip:
  Increase the border size if the page is not fully within the default region, or decrease it to avoid detecting spurious edges from content near the image boundaries.

Can be one of the following types:

The 'Angle Precision' property sets the angular increment (in degrees) for each iteration of edge angle detection. Smaller values increase the precision of edge detection but also increase processing time, as more angles are tested.

• Performance vs. Accuracy:
  Lower values (e.g., 0.125 or 0.25) provide finer detection and are recommended for pages with subtle skew or perspective. Higher values (e.g., 1.0) speed up processing but may miss small angle variations.
• Example:
  With a value of 0.125, the algorithm tests every 1/8th of a degree. For most documents, 0.125 to 0.25 is a good starting point.
• Tuning Tip:
  If edge detection is missing or imprecise, decrease the angle precision. If processing is slow and results are already accurate, consider increasing it.

The 'Threshold' property defines the minimum strength required for a detected line to be considered a valid page edge. This value is expressed as a percentage of the image width or height.

• Sensitivity:
  Higher values require longer or more continuous lines, reducing the chance of false positives from noise or short marks. Lower values make the algorithm more sensitive, which can help with faint or broken edges but may increase false detections.
• Example:
  If set to 0.75 (75%), a vertical line in the left border region must be at least 75% the height of the image to be considered an edge.
• Diagnostics:
  Use the 'Edges' diagnostic image to review which lines are detected and adjust the threshold accordingly.
• Tuning Tip:
  For strong, clear edges, use a higher threshold (e.g., 0.8). For faint or partial edges, lower the threshold (e.g., 0.5).

Can be one of the following values:

The 'Interpolation Mode' property determines the algorithm used to resample pixels during the warp (de-warping) operation. This affects both the visual quality and processing speed of the extracted page image.

• Cubic:
  Produces smooth, high-quality results and is recommended for most documents, especially those containing text.
• NearestNeighbor:
  Fastest, but may introduce jagged edges or pixelation. Suitable for binary images or when speed is critical.
• Other Modes:
  Additional modes may be available for specific use cases. Review output images to ensure text and graphics remain sharp.
• Tuning Tip:
  Use 'Cubic' for best quality unless performance is a concern or the image is already binary.

The 'Apply Edge Smoothing' property controls whether additional smoothing is applied to the edges of the extracted page during the warp operation.

• Purpose:
  Smoothing can reduce jagged or pixelated edges, especially when the page is extracted from a low-resolution or noisy image.
• When to Enable:
  Enable edge smoothing if the output image shows visible stair-stepping or roughness along the page borders.
• Performance:
  Enabling smoothing may slightly increase processing time, but the impact is minimal for most images.
• Tuning Tip:
  If the extracted page looks clean and sharp, you may disable smoothing for maximum speed.

The 'Command Type Name' property shows the user-friendly name of the IP Command as it appears in the Grooper UI. This helps users identify which command is being configured or executed, especially when working with IP Profiles containing multiple commands.

This property is read-only and is automatically set based on the command's type.

The 'Supported Pixel Formats' property lists the image pixel formats that the IP Command can process. If an image does not match one of these formats, the command will be skipped for that image.

• By default, all pixel formats are supported unless a derived command restricts this list.
• Example formats include Pixel8bppGrayscale, Pixel24bppBgr, and Pixel1bppIndexed.

Use this property to verify compatibility when building IP Profiles or troubleshooting why a command did not execute.