A new technique for edge detection of chromosome G-BAND images for segmentation

Abstract

Chromosome edge detection of G-band type images are an important for segmentation. Generally, the chromosome G-band images consisted of the noise, lack of contrast and hole in the images. The chromosome edge can mislead of the edge detection method, particularly the chromosome overlaps and the chromosome touches. It’s difficult a clear edge of the chromosome edge images. The edge detection method is difficult when the noise appear. This paper proposed approach the chromosome edge detection in the chromosome segmentation system. The chromosome edge detection method applied the FloodFill, Erosion and Canny based on the chromosome G-band images. The experimental results give the best performance for chromosome segmentation system. A success rate of proposed method achieved 98.43%.