EFFICIENT MEDICAL IMAGE SEGMENTATION USING FUZZIFICATION OF ACTIVE CONTOURS WITH SRGA
Kavitha Ananth, S. Pannirselvam · 2012
Many image processing techniques have been developed over the past two decades to help users in diagnosing medical images. Segmentation is a significant feature of medical image processing, where clustering system is broadly employed in biomedical applications mainly for disease diagnosis recognition. In existing approach, image segmentation is done through multi-resolution stochastic level set method (MSLSM), but the downside of MSLSM is that if topology changes occur, it presented nonparametric topology-constrained segmentation model. To improve the image segmentation more effective, our first work planned to do image segmentation with level set method with geodesic active contour to examine medical image disease diagnosis. Then, cluster object configuration is ended with fuzzification of increasing active contours with the earlier recognized contours of diseased image portions. Fuzzy active functions are generated with detected contours by adapting the fuzzy relevance feed back mechanism. After detecting the contour with the Fuzzy relevance feedback mechanism, the detected contours are segmented based on its contour growth in a cluster group of medical image segment portions. To enhance the medical image segmentation, in this work, we plan to present a new technique fuzzifcation of active contours with SRGA (Seeded Region Growing Algorithm) which group the contour based on its growth in a segmented medical image level. With segment portions new similar regions can be traced out with automatic seeded growing method. After separating the diverse disease diagnosed regions, we use our proposed Seeded Region Growing Algorithm (SRGA) to isolate standard and irregular regions in the medical image. An experimental evaluation is conducted with bench mark data sets obtained from UCI repository to estimate the performance of the proposed efficient medical image segmentation using cluster based fuzzification of active contours with SRGA (EMISCF) and parametric evaluations are measured in terms of Contour growth, segment size, efficiency, error rate.