ImageGear for C and C++ on Windows v19.10 - Updated
IG_REC_image_despeckle
API Reference Guide > Recognition Component API Reference > Recognition Component Functions Reference > Recognition Image Preprocessing Functions > IG_REC_image_despeckle

Performs an immediate adaptive noise removal on the bi-tonal image specified, irrespective of its resolution.

Declaration:

 
Copy Code
AT_ERRCOUNT ACCUAPI IG_REC_image_despeckle(
   HIG_REC_IMAGE hImage
);

Arguments:

Name Type Description
hImage HIG_REC_IMAGE Handle of the image.

Return Value:

Returns the number of ImageGear errors that occurred during this function call.

Supported Raster Image Formats:

Indexed RGB - 1 bpp.

Example:

 
Copy Code
AT_ERRCOUNT ErrCount = 0;
HIGEAR higImage = 0;
HIG_REC_IMAGE higRecImage = 0;

ErrCount += IG_load_file("Image.tif", &higImage); 
ErrCount += IG_REC_image_import(higImage, &higRecImage);
ErrCount += IG_REC_image_despeckle(higRecImage);

//...

ErrCount += IG_REC_image_delete(higRecImage);
ErrCount += IG_image_delete(higImage);

Remarks:

  1. An image despeckled by this function replaces the original one and is available for display and export.
  2. If the image specified is not a bi-tonal one, the function sets ImageGear error with Value1 set to IGE_REC_EXT_IMG_BITSPERPIXEL_ERR.
  3. This function is useful for typical images that originally had a grayed or colored background that was not completely removed. It can be useful also for faxed pages or those with a spotted background.
  4. The algorithm will run whatever the resolution, however, it is optimized for images with at least 280 dpi resolution.
  5. As an alternative, the setting IG_REC_despeckle_mode_set is available. While it is enabled, all bi-tonal images with a resolution of 280 dpi or higher are copied and despeckled whenever a function is called which can benefit from a cleaner image, i.e., pre-processing, auto-zoning, or recognition. However, these despeckled images are not available to the application.
  6. This function is not applied to grayscale or color images, because any implicit conversion the recognition engine makes internally results in cleaned (but hidden) bi-tonal files.

See Also

HIG_REC_IMAGE