Novel Approach for Discovery of Edges in Medical Images

Edge may be characterized as boundary between an object and background. Edge detection is a classical problem in the field of computer vision and image processing. It is crucial to understand the mechanism of Edge Detection as this is the forefront for the development of applications in domains like tumor detection, object detection and other medical imaging applications.. Edge detection is one of the starting steps for number of operations in medical image analysis where it can be utilized to detect irregularities like tumors, disjoints etc. The goal of the edge detection process in a digital image is to determine the frontiers of all represented objects by automatically processing color or gray level information in each pixel. The image intensity shows sudden changes at edges. Edge detection usually involves the calculation of derivative of the image intensity so that relative change can be measured and if magnitude of this change is high pixel is considered as edge pixel. Purpose of all edge detection algorithms is to detect and highlight these discontinuities. This paper presents a novel approach for the detection of these discontinuities in medical images. This paper is organized as follows: after this introductory part, section II focuses on some of the work already carried out in field of edge detection, section III introduces the proposed method, results are included in next section followed by conclusion and future directions.