Segmentation of Medical Images Based on LS-SVM using Low Level Features

The amount of medical digital images that are produced in hospitals is increasing incredibly so, the need for systems that can provide efficient segmentation and retrieval of images of particular interest is becoming very high. Image segmentation partitions an image into non overlapping regions, which ideally should be meaningful for a certain purpose. In recent years, many image segmentation algorithms have been developed, but they are often very complex and some undesired results occur frequently. In this paper, we present an effective color image segmentation approach based on pixel classification with modified least squares support vector machine (LS-SVM). In this technique, firstly the images are segmented using proposed LS-SVM approach and then its results are compared with performance matrices. This image segmentation not only can fully take advantage of the local information of color image, but also the ability of LS-SVM classifier and removes the problem of over segmentation. Experimental evidence shows that the proposed method has very effective segmentation results and computational behavior, and decreases the time, increases the quality of color image segmentation and eliminates the problem of over segmentation in comparison with the state-of-the-art segmentation methods recently proposed in the literature.