Quality Evaluation for Edge Detection of Chromosome G-band Images for Segmentation

Abstract

Edge detection of chromosome in G-band type image is an important preprocessing step in segmentation. A chromosome type G-band image is generally very noise and poor image quality, lack of contrast and hold in image. A lot of edge caused by chromosome can easily mislead the edge detection algorithm. Particularly, the chromosome overlapping and touching, then it is very difficult to get a clear edge of the femur head. These extraneous edge and noise make edge detection very difficult and challenging, which is not well solved. This paper presents analysis of evaluation chromosome G-band image edge detection. It has been appeared in literature that there are 4 different techniques, i.e. Canny, Laplacain, Robert's, and Sobel, on chromosome image type G-band. From our study, we found that Robert's give promising results in edge detection. A success rate of 97.69% has been achieved.