A Fast Watersheds Approach for Image Segmentation Based on Relative Potential Energy
Traditional watersheds methods face the difficulty of “over-segmentation for images with many gray details and noises. In this paper, the traditional segmentation method is improved. Firstly, “Opening and Closing in morphology is used to smooth image, and some unnecessary details are removed. Then, this paper uses the fast watersheds approach for image segmentation based on relative potential energy method. The original potential energy is replaced by relative potential energy. This approach gains higher efficiency and less run time. The experiments show a remarkable reduction of segmentation regions. This approach show an improvement on the segmentation performance and it can remove the unimportant details and noise in the image to extract the clearly whole figure of the image fleetly.
fast watersheds transformation relative potential energy picture segmentation
Lili Zhang Xiaojun Zhang Zhijing Liu
the Research Center of Computer Information Applications, Xidian University, Xi’an 710071, China Xidian University, Xi’an 710071, China
国际会议
武汉
英文
2007-09-21（万方平台首次上网日期，不代表论文的发表时间）