Based On Fuzzy Clustering Of FLICM And Level Set For Segmentation Of Medical Images

Image segmentation refers to separate different areas of the region which has a special meaning in the image,the medical image segmentation refers to the image is divided into disjoint tissues or organs accodring to the pixels of correlation of pixels in image.Compared to ordinary images,medical images have the characteristic of low contrast more noise and blurred boundaries between different tissues or lession,besides medical image itself is complex,the use of a conventional automatic segmentation method causes the vast most of the segmentation algorithm segmentation results are not satisfactory,the segmentation speed and performance needs to be improved,so the medical image segmentation has been a hot and difficult experts and scholars in the field of image processing.In this paper,the use of fuzzy clustering and level set algorithm for medical image segmentation,the traditional FCM algorithm,the images have a good segmentation results without noise,while low SNR image segmentation is not ideal due to ignoring the pixels spatial location information,for this shortcoming,this paper describes some Improved classic algorithms,including FCMS,FCMS1,FCMS2,En FCM,FGFCM and so on,compare with traditional FCM algorithm,the result of segmentation has been greatly improved by the improved algorithm,while the parameters need to be setted,what’s more the size of these parameters directly affect the quality of segmentation,and adjusting the parameters is very difficult.So we use a FLICM algorithm that takes into account not only the spatial information between pixels,and does not contain the necessary parameters in addition to m,C of other parameters.When using the level set algorithm for image segmentation,not only consider Spatial information of the image but also the boundary information of the image,it is often able to achieve good segmentation results,but the traditional level set algorithm needs to continue to initialize,for this shortcoming,this paper present an improved algorithm to overcome the problem of computational intense and require periodic initialization,but the algorithm still exists some problem of requiring the initial outline be given manually and how to chose the control parameters optimally.Considering these factors,we use the method which combine FLICM algorithm with Level Set in the paper.Firstly,we adopt FLICM algorithm to pre-segment medial image,an initial outline can be obtained,then we adopt Level Set algorithm to segment image meticulously by the intial outline.Not only the initial outline and without being provided or artificially but also the Level set parametersin without setted artificially in the new algorithm,So the new algorithm can achieve automatic image segmentation.And through the experiment,we can know this algorithm have some advantage which are Better noise immunity,fast segmentation speed, and high precision segmentation.