A Medical Image Segmentation System Design Based On The Semi-supervised Fuzzy Clustering

Image segmentation is an important issue in the field of image processing and computer vision, medical image segmentation is an important application in the field of image segmentation, Since the medical image in the imaging process by a variety of factors (such as resolution, illumination conditions, etc.), resulting in having a certain degree of uncertainty, the similarity between the target and the background to be extracted, for this uncertainty good solution is to use the fuzzy image processing technology. Fuzzy clustering method is an important theoretical branch which can deal with the problem of image segmentation. The fuzzy C-means clustering (Fuzzy C-means, FCM) algorithm has been widely used in practical applications, clustering it boils down to a nonlinear programming problem with constraints solved through optimization of the objective function to achieve data set fuzzy divided. Semi-supervised learning is the use of a small amount of labeled data and unlabeled data for training and classification,Semi-supervised learning in natural language processing, feature recognition, text categorization, retrieval and medical image analysis system has been widely used.On the basis of in-depth study on the basic theory of semi-supervised fuzzy clustering and semi-supervised learning method for the existing problems in the traditional algorithms,also the practical difficulties in the field of medical image segmentation, an improved semi-supervised fuzzy clustering class of algorithms, and has been applied in the human brain MRI (Magnetic Resonance) image segmentation. In this paper, the following research:(1) Based on the traditional fuzzy C-means clustering, a semi-supervised fuzzy clustering was put forward.Experimental results on UCI benchmark data (Iris and Wine) also show the effectiveness of the algorithm, which significantly improved semi-supervised fuzzy clustering algorithm clustering accuracy than the original fuzzy C-means clustering algorithm.(2) In this paper, semi-supervised fuzzy C-means clustering in the case of sparse marker point will become the problem of the traditional fuzzy C-means clustering, and then an improved semi-supervised fuzzy C-means clustering algorithm has been put forward. This article will improve the semi-supervised fuzzy clustering algorithm applied to the medical image segmentation and clustering effect a comprehensive evaluation of segmentation accuracy and speed indicators. (3) By using the GUI graphical user interface,a system for medical image segmentation was designed, which can visually read and segment the medical images by setting different parameters.it also provides a good experimental platform for subsequent theory and algorithm research.