MR Brain Image Segmentation Using an Improved Kernel Fuzzy Local Information C-Means Based Wavelet, Particle Swarm Optimization (PSO) Initialization and Outlier Rejection with Level Set Methods

This paper, presents a new image segmentation method based on Wavelets, Particle Swarm Optimization (PSO) and outlier rejection caused by the membership function of the Kernel Fuzzy Local Information C-Means (KFLICM) algorithm combined with level set is proposed. The segmentation of Magnetic Resonance (MR) images plays an important role in the computer-aided diagnosis and clinical research, but the traditional approach which is the Fuzzy C-Means (FCM) clustering algorithm is sensitive to the outlier and does not integrate the spatial information in its membership function. Thus the algorithm is very sensitive to noise and in-homogeneities in the image, moreover, it depends on cluster centers initialization. A novel approach, named improved IKFLICMOR is presented to improve the outlier rejection and reduce the noise sensitivity of conventional FCM clustering algorithm. To get the first image segmentation, the traditional FCM is applied to low-resolution image after wavelet decomposition. In general, the FCM algorithm chooses the initial cluster centers randomly, but the use of PSO algorithm gives us a good result for these centers. Our algorithm is also completed by adding into the standard FCM algorithm the spatial neighborhood information. These a priori are used in the cost function to be optimized. The resulting fuzzy clustering is used as the initial level set function. The results confirm the effectiveness of the IKFLICMOR associated with level set for MR image segmentation.