Application Of Improved FCM Algorithm In Brain Image Segmentation

Brain image segmentation based on MRI images is an accurate extraction of brain regions from MRI images,and is great significant to diagnose of brain diseases.Therefore,it is necessary to study the segmentation of brain MRI image and has certain practical significance.Fuzzy C-Means clustering algorithm is the most suitable method applied to MRI segmentation of brain.The essence of using FCM to improve the FCM algorithm is to control the membership of the center pixel by the distance between the neighborhood pixel and the cluster center.In this paper,FCM and its improved algorithm are studied in detail,and it is found that,determining the degree of influence of the neighborhood pixels on the central pixel is the key to the success of algorithm improvement in this process.The existing improved algorithm has the following disadvantages:the one is that the influence factors are unified and the difference between pixels is neglected,to lead to the inaccuracy of the segmentation result.The one is that the segmemtation completely dependent on the distribution of the image itself can obtained the better effect of segmentation,but it is very sensitive to noise,and has great influence on the subsequent intracranial lesions within the larger regional division.In view of the above situations,this paper introduces pixel correlation based on pixel gray level difference,to determine the influence degree of neighborhood pixel on the central pixel.On this basis,a new FCM improved algorithm based on neighborhood pixel correlation is proposed.The algorithm works out the influence degree of the neighborhood pixels on the central pixel by the correlation of the gray-level difference between the domain pixel and the center pixel.Then,the distance between the neighborhood pixel and the cluster center is used to control the membership of the center pixel.In this paper,a post-segmentation strategy is also presented to reduce the influence of noise and optimize the segmentation result.At last,this algorithm is implemented by MATLAB tool,and compared with FCM,FCMS,FCMS1 and FLICM algorithms.The feasibility of the presented algorithm and the accuracy of the segmentation result are verified by evaluating the algorithm and the experimental results according to the relevant evaluation criteria.