A Proposed Hybrid Fuzzy C-means Algorithm With Cluster Center Estimation For Leukemia Image Segmentation

Fuzzy C-means clustering (FCM) is an important technique used in cluster analysis. The standard FCM algorithm calls the centroids to be randomly initialized resulting in the requirement of making estimations from expert users to determine the number of clusters. To overcome these observed limitations of applying the FCM algorithm, an efficient image segmentation model, Hybrid Fuzzy C-means Algorithm with Cluster Center Estimation (HFCMCCE) using subtractive clustering for Leukemia infected blood sample Image Segmentation is presented in this paper. In this algorithm, the image is initially subjected to Partial Contrast Stretching (PCS) method to modify the force level of the dark scale image, then Subtractive Clustering Method (SCM) is applied to determine the thickness measure of the pixel and the pixel with the highest thickness measure is marked as the first cluster center. These cluster centers are used in FCM to perform the image segmentation to obtain the required segmented image. The performance of this algorithm is then compared with other existing algorithms, on the basis of image quality measures such as PSNR (Peak-Signal-to-Noise ratio), MSE (Mean Square Error).