Abstract :
Segmentation techniques play an important role in medical image analysis. More research is required to
understand the performance of various segmentation techniques on microscopic medical images. In this paper, four
segmentation techniques are analyzed to segment the chromosome images. Performance metrics like SSIM, MSSIM
are evaluated for Fuzzy C-Means of clustering, K-Means clustering, Kernel weighted Fuzzy C-Means (KWFCM)
and Watershed segmentation. These techniques are compared subjectively and objectively. KWFCM is found to be
good and suitable for segmentation of chromosomes from the simulation results.
Keyword :
FCM, K-Means, KWFCM, Watershed.