Grooper Help - Version 25.0
25.0.0017 2,127
  • Overview
  • Help Status

Binary Dropout

Feature Dropout Grooper.IP

Serves as the base class for feature dropout commands that perform detection on a binarized (black and white) image.

Remarks

The Binary Dropout class provides a foundation for feature dropout commands in Grooper that require a binarized version of the input image to accurately detect and remove unwanted features. Derived commands use this binarized image to identify regions such as barcodes, halftones, lines, or other visual artifacts that should be masked or suppressed in the final output.

How Binary Dropout Works

  1. The input image is converted to black and white using the settings defined in the 'Binarization Settings' property.
  2. Feature detection is performed on the binarized image by the derived command, which generates a mask identifying regions to be removed.
  3. The mask is applied to the original image using the configured dropout method, resulting in an output image with the targeted features suppressed.

Configuration and Usage

  • Use the 'Binarization Settings' property to control how the input image is binarized. Fine-tuning these settings is critical for accurate feature detection, especially on documents with variable quality, contrast, or background noise.
  • Derived commands may expose additional properties to further control feature detection and mask generation.
  • The binarized image is used only for detection; the final output is based on the original image with the mask applied.

Supported Pixel Formats

All common pixel formats are supported, including Pixel8bppGrayscale, Pixel24bppBgr, and Pixel1bppIndexed. Images are automatically converted as needed for binarization and feature detection.

Diagnostics

When run in diagnostic mode, Binary Dropout outputs a diagnostic image named Binarized that shows the result of the binarization process. This allows you to review and tune the binarization output to ensure features are clearly visible for detection and that only the intended regions are masked.

Notes

  • The effectiveness of feature dropout depends heavily on the quality of the binarized image. Review the diagnostic output to ensure that the binarization settings are appropriate for your documents and that only the intended features are being detected and removed.
  • Binary Dropout is not used directly, but provides essential infrastructure for specialized dropout commands such as Barcode Removal and others.

Properties

NameTypeDescription
General
Image Preprocessing
Command Info

Derived Types

There are 6 implementations of Binary Dropout.

Barcode Removal Removes barcode regions from an image using configurable detection and masking options.
Blob Removal Removes blobs from the image that meet specific size, shape, or fill requirements.
Border Fill Removes artifacts and unwanted features near the edges of binary images.
Halftone Removal Removes halftone regions from an image to improve OCR and document clarity.
Hole Punch Removal Removes circular hole punch artifacts from an image, typically found along the edges of scanned documents.
Speck Removal Removes small specks from an image.

See Also

Used By

Notification