Segmentation of Masses in Mammographic Images

Image Processing is a process of performing few operations on an image either to obtain an enhanced image or to extract useful information from it. In image processing input provided is an image and output can be image or features associated with particular image. Image processing is one of the most rapidly growing technology in computer science field which also forms the core research area. Image processing techniques are widely used in detection of breast cancers. Segmentation is a process of dividing an image into distinct regions with each pixel having similar properties. Region should strongly relate to the feature of interest to make it more meaningful and useful image analysis. Segmentation is the first step in transforming a greyscale image into one of more other images i. e. from low level image processing to high level image processing. The success of image analysis depends on segmentation but partitioning of an image accurately is a challenging task. This paper presents an approach of automatic breast mass segmentation which has three stages: contour initialization, construction of fuzzy contour and estimation of fuzzy membership maps.