A hybrid algorithm for segmentation of range images

The authors present a technique for the segmentation of range images. It tends to be a hybrid approach that combines edge detection and region growing techniques for accomplishing the task of segmentation. The surface coherency property of range data is an important attribute of the technique. The cross-breeding approach produces a segmentation map and an edge map. The incorporation of the region growing process eliminates internal micro edges and produces a segmentation map. These two maps would undoubtedly be of great value for the higher-level recognition process. Experimental results show that the edges detected and the regions partitioned tend to have a good localization property