An Efficient Clustering Algorithm for MRI Brain Image Segmentation

Image processing plays an important role in medical field because of its capability. Particularly, image segmentation offer several guides in medical field for analyzing the captured image. Usually, the medical images are captured via different medical image acquisition techniques. The captured image may be affected by noise because of some faults in the capturing devise; this will leads to false diagnosis. This paper focuses on enhancing the captured brain image by using image segmentation technique. Usually, brain image is captured using Magnetic Resonance Imaging (MRI) technique. The captured brain image will have high amount of noise or distortion, this noise must be removed before it is used for diagnosis purpose. Brain segmentation is widely applied for removing those noises to produce the clear image. The segmentation can be achieved with the help of clustering techniques. The widely used clustering technique for brain image segmentation technique is Fuzzy C-Means (FCM) clustering. But FCM will result in poor segmentation when more edge regions are involved. To overcome this problem, Fuzzy Possibilistic C-Means Algorithm (FPCM) is introduces. Even FPCM will result in poor segmentation when more noise are involved. To overcome all these problems, a Modified Fuzzy Possibilistic C-Means Algorithm (MFPCM) is proposed in this paper. The experimental result show that the segmentation resulted for the proposed technique is better when compared to the existing methods.