Image Segmentation based on Markov Random Field Probabilistic Approach

Image segmentation is the process of partitioning the digital image into many quarters or pixels. It is very important to build a mathematical model to segment out the area of interest than using complex techniques. Markov Random Field (MRF) is a tool which is reducing the tough procedure as well as simplifies the procedure of the handling techniques. The aim of this paper is to segment cancer affected region of the input mammogram image using Markov Random Fields probabilistic measure. The goal of this technique is to simplify as well as to give a meaningful representation of the image. The pectoral muscle always mimics as a tumor is identified from the actual breast region. As an initiative, the input image is processed to highlight the upper region muscle using enhancement and adaptive thresholding. The pixel variation in muscle density is differentiated and is marked. The boundary of the chest muscle is highlighted. The remaining breast density is subjected to Clique potential with Markov to label cancer affected regions and is segmented.