Medical Image Segmentation Based on Watershed Transformation and Rough Sets
Traditional watershed algorithm often causes over-segmentationbecause of its high sensitivity to the weak edge and the noise. To overcome this drawback and in light of thecharacteristics of medical image, a new segmentation algorithmbased on watershed transformation and rough set theory isproposed. The original image is partitioned into the edge-detailsub-image and smooth sub-image according to indiscernibilityrelation of rough set theory. Two enhancement methods aredesigned for the two sub-images, and watershed transformationis used for the further segmentation in the smooth sub-image.Finally, combine the two processed sub-images to obtain thesegmentation result. The proposed algorithm has been executedon Magnetic Resonance Imaging (MRI) image, the analysis ofcompare between conventional watershed algorithm and theproposed algorithm is given. The experimental result shows thatthis method is efficient to restrain the over-segmentation, thusobtaining good segmentation results.
Ran Li
Dept.of Electronic and Communication Engineering North China Electric Power University Baoding, China
国际会议
成都
英文
1-5
2010-06-18（万方平台首次上网日期，不代表论文的发表时间）