Description: Edge detection is important in image procession. This paper made a summary and analysis of edge detecting algorithm and edge detector. Canny has proposed three continuous criteria to compare the performance of different filters: good detection, good localization and low-responses. Based on these criteria he got optimal filter for edge detection: derivative of Gaussian function. After more than ten years research, Canny’s theory has been ameliorated in many aspects, this paper also made a review of it. Based on the practice and theory. Demigny gave three discrete criteria for edge detection like Canny’s criteria and he has proofed that the third criterion can be replaced by an appropriate thresholding operation. This paper used numerical method to get the optimal filter and smooth operator under the Demigny’s criteria. Then I combine these filters and Canny’s edge detecting technique to get an integrated edge detecting algorithm. I have implemented the algorithm using VC++. From the res
File list (Check if you may need any files):
edgedetection2003.pdf
Download