Segmentation methods can also be applied to edges obtained from edge detectors. Lindeberg and Li  developed an integrated method that segments edges into straight and curved edge segments for parts-based object recognition, based on a minimum description length (M DL ) criterion that was optimized by a split-and-merge-like method with candidate breakpoints obtained from complementary junction cues to obtain more likely points at which to consider partitions into different segments.
The following structuring elements can be used for 2D images:
The most fundamental class we're about discuss at this point is ImageSegmentation class implementation. In the following class we normally declare the method public void Compute( string InputFile, string OutputFile) which is the main of the image segmentation engine that performs the actual k-mean clustering. Also, besides of this method, we'll implement two more simple methods that perform the Euclidian distance computation, which we'll discuss more about later.
MorphoLibJ is a collection of mathematical morphology methods and plugins for ImageJ, created at INRA-IJPB Modeling and Digital Imaging lab.
Huchuan Lu’s publication(Selected) DUTS Saliency Dataset Relseased [ Website ]