Based On Improved Canny Algorithm For Image Edge Detection

With the rapid development of computer technology, image edge detection technique has become the field of image processing and computer vision research focus, digital image edge detection is image segmentation, image recognition, image analysis such as area shape extracting the important foundation. The main contents are as follows:Firstly, introduces the basic concepts of digital image processing and its applications, an overview of background to the study of edge detection significance and basic concepts, Secondly, respectively introduces classic image edge detection operators, through theoretical analysis and experimental results compare the advantages and disadvantages of each operator, again, discusses the Canny criterion and Canny edge detection algorithm, describes in detail Canny edge detection algorithm realization of process, and analyzes the advantages and disadvantages of traditional Canny algorithm, Finally, the traditional Canny algorithm adopts the gaussian filter and its gaussian parameter uncertainty caused images appear untrue edge and edge details is lost and other defects, this paper proposes a method based on median filter improvement the Canny algorithm, and improve gradient amplitude calculation method, analyzes the realization process improvement method, and the noise in image different simulation experiments and use the peak value signal-to-noise ratio (PSNR) to compare the two algorithm's anti-noise sex, based on the analysis of experimental results available, with the median filter instead of the gaussian filter improved algorithm has higher signal-to-noise ratio, it can extract accurate edge information, effectively overcome traditional Canny Gaussian filtering algorithm is smooth and lost some dispersion-decreasing edge, it can get good edge detection effect.