RecursiveBlobCounter Class

The class counts and extracts stand alone objects in images using recursive version of connected components labeling algorithm.

Note
The algorithm treats all pixels with values less or equal to BackgroundThreshold as background, but pixels with higher values are treated as objects' pixels.

Note
Since this algorithm is based on recursion, it is required to be careful with its application to big images with big blobs, because in this case recursion will require big stack size and may lead to stack overflow. The recursive version may be applied (and may be even faster than BlobCounter) to an image with small blobs - "star sky" image (or small cells, for example, etc).

For blobs' searching the class supports 8 bpp indexed grayscale images and 24/32 bpp color images. See documentation about BlobCounterBase for information about which pixel formats are supported for extraction of blobs.

Sample usage:

// create an instance of blob counter algorithm
var bc = new RecursiveBlobCounter();

// process binary image
bc.ProcessImage(image);

// process blobs
foreach (Rectangle rect in bc.GetObjectsRectangles())
{
    // ...
}

Reference