Title :
Range image segmentation using mathematical morphology
Author :
Deng Shiwei ; Yuan Baozong
Author_Institution :
Inst. of Inf. Sci., Northern Jiaotong Univ., Beijing, China
Abstract :
We introduce a range image segmentation method using simple morphological operators. The paper describes how to obtain peak and valley image using edge-strength operator, peak and valley detection operators. This procedure makes the depth discontinuities contained in the peaks and the orientation discontinuities in the valleys. The depth discontinuities are extracted and the segmentation is accomplished by controlled region growing. Several examples of the experimental results from real range images are presented. The experimental results prove this range image segmentation technique quick and robust.<>
Keywords :
feature extraction; image segmentation; mathematical morphology; controlled region growing; depth discontinuities; edge-strength operator; mathematical morphology; orientation discontinuities; peak and valley image; range image segmentation; real range images; simple morphological operators; Area measurement; Computer vision; Image edge detection; Image segmentation; Information science; Labeling; Polynomials; Robustness; Shape measurement; Surface morphology;
Conference_Titel :
TENCON '93. Proceedings. Computer, Communication, Control and Power Engineering.1993 IEEE Region 10 Conference on
Conference_Location :
Beijing, China
Print_ISBN :
0-7803-1233-3
DOI :
10.1109/TENCON.1993.320184