Hybrid Segmentation Method for Malignancy Detection Using Fuzzy-C-Means and Active Contour Model
S. S. Sarmila, S. Naganadini Sujatha · 2012
Image segmentation is the first stage of processing in many practical computer vision systems. Development of segmentation algorithms has attracted considerable research interest, relatively little work has been published on the subject of their evaluation. Hence this paper enumerates and reviews mainly the image segmentation algorithms namely Fuzzy C means, Active Contour Model. The rapid progress in computerized medical image reconstruction, and the associated developments in analysis methods and computer-aided diagnosis, has propelled medical imaging into one of the most important sub-fields in scientific imaging. The proposed paper is an innovative frame work of hybrid segmentation technology with region based techniques and active contour models. The drawback of the parametric Active contour models is manual control points. But the proposed technology automatically assigned control points using Fuzzy-c-means clustering techniques. This approach is suitable for medical application like cancer cells for MRI, PET scan. Since the medical images are in very low contrast. Fuzzy-C-means gives the approximated boundary from that the control points are randomly selected. This combined approach will give accurate result for especially cancer cell detection. The FCM(3) performed the skin detection based on decision rules in hybrid space. (3) a novel segmentation technique for color images is presented. The segments in images are found automatically based on a novel FCM algorithm by using I and H components in HIS color space to form a new feature. (4) SWFCM cluster algorithm and we called it NSWFCM. The system can be used as a primary tool to segment unknown colour images. The algorithm has been implemented on a set of images. Results show that the system performance is robust to different types of images compared to FCM and SWFCM. The difficulty over re- initialization process is handled by Chumming Li et, al. In this paper level set method without re-initialization procedure was introduced. Here the new variation AL formulation of level set function with a closed sighed distance function is used to eliminate the costly re-initialization process. clustering algorithms namely centroid based K-Means and representative object based Fuzzy C-Means. These two algorithms are implemented and the performance is analyzed based on their clustering result quality. The behavior of both the algorithms depends on the number of data points as well as on the number of clusters (5).points are generated by two ways, one by