Feature Dropout

Provides a base class for image processing commands that remove or suppress specific features from an image using a mask-based approach.

Remarks

The Feature Dropout class serves as the foundation for all feature dropout commands in Grooper. These commands are designed to identify and remove targeted regions or features from an image, such as halftone patterns, lines, or other unwanted elements, by generating a mask and applying a configurable dropout method.

Feature dropout commands are typically used to clean up scanned documents, improve OCR results, or prepare images for further processing by eliminating visual noise or artifacts.

How Feature Dropout Works

1. The command analyzes the input image and generates a mask that identifies the regions to be removed.
2. The selected 'Dropout Method' determines how the masked regions are removed or replaced in the output image.
3. The resulting image has the specified features suppressed or eliminated, while preserving the rest of the content.

The specific logic for generating the mask is implemented by derived classes, each targeting a different type of feature (e.g., halftone removal, line removal).

Supported Pixel Formats

All pixel formats are supported by Feature Dropout commands, unless a derived command restricts this list. This includes common formats such as Pixel8bppGrayscale, Pixel24bppBgr, and Pixel1bppIndexed.

Diagnostics

When run in Grooper testing interfaces (diagnostic mode), Feature Dropout commands generate additional output to assist with configuration and troubleshooting. This may include:

• Diagnostic images showing the generated mask overlaid on the original image (in red).
• Log messages indicating the timing and results of the mask application.

These diagnostics can be used to verify that the mask is correctly identifying the intended features and to fine-tune command parameters for optimal results.

Configuration and Usage

To use a Feature Dropout command, add it to an IP Profile or configure it as a property of a Grooper object. Set the 'Dropout Method' property to control how masked regions are removed. Additional properties may be available in derived commands to further customize feature detection and removal.

For best results, review the diagnostic output to ensure that only the desired features are being dropped out, and adjust configuration as needed.

Properties

Name	Type		Description
General
Dropout Method	Dropout Method	►	Specifies how masked regions will be removed from the image. ConverterType Selector Requiredtrue Can be one of the following types: Value Description Fill Fills masked regions of an image with a solid color. Inpaint Digitally restores masked regions of an image by filling in missing or damaged areas. The 'Dropout Method' property determines the technique used to remove or replace regions of the image identified by the feature mask. This allows you to control the visual result of the dropout operation, depending on your processing goals. Common dropout methods include: Filling masked regions with a solid color (such as white or the detected background color). Replacing masked regions with interpolated content from surrounding pixels. Applying custom fill logic as defined by the selected Dropout Method. Choosing the appropriate dropout method is important for maintaining the readability and appearance of the processed image. For example, filling with white may be suitable for removing background noise, while interpolation may be preferred when removing lines or halftones from text regions. Example: To remove halftone patterns and fill the resulting regions with the detected background color, select the 'Background Fill' option from the 'Dropout Method' dropdown. Review the diagnostic images generated in diagnostic mode to verify that the selected method produces the desired effect on your sample images.
Command Info
Command Type	String	►	The display name of this image processing command. 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.
Supported Pixel Formats	String	►	The pixel formats supported by this command. 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.

Derived Types

There are 8 implementations of Feature 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.
Scratch Removal	Digitally removes scratches occurring on film-based media such as microfilm, microfiche, aperture cards, or 35mm slides.
Shape Removal	Detects and removes regions of an image which match one or more trained image templates.
Speck Removal	Removes small specks from an image.

See Also

FillInpaint

Used By

IP Step