A Deterministic Edge Detection Using Statistical Approach

A large number of edge detectors are available in image processing literature where the choices of input parameters are to be made by the user and are made on an informal basis. In this paper, an edge detector is proposed, where thresholding is performed using statistical principles. Local thresholding of each individual pixel which depends upon the statistical variability of the gradient vector at that pixel is made. Such a standardization statistic based on the gradient vector at each pixel is used to determine the eligibility of the pixel to be an edge pixel. The results obtained from the proposed method are found to be comparable to those from well-known edge detectors. However, the values of the input parameters providing the appreciable results in the proposed detector are found to be more established than other edge detectors and possess statistical elucidation. The results obtained from the proposed algorithm are compared with Canny´s edge detector which is more popular among different edge detectors. The proposed algorithm is implemented using MATLAB-7.1.