This section describes functions performing flood filling of connected areas. Flood filling means that a group of connected pixels with close values is filled with, or is set to, a certain value. The flood filling process starts with a specified point (“seed”) and continues until it reaches the image ROI boundary or cannot find any new pixels to fill due to a large difference in pixel values. For every pixel filled, the functions analyze neighbor pixels:
- 4 neighbors (except diagonal neighbors); this kind of connectivity is called 4-connectivity and the corresponding function name includes 4Con, or
- 8 neighbors (diagonal neighbors included); this kind of connectivity is called 8-connectivity and the corresponding function name includes 8Con.
These functions can be used for:
- segmenting a grayscale image into a set of uni-color areas,
- marking each connected component with individual color for bi-level images.