The Algorithm Research Of Gray Image Edge Detection

Edge detection plays a crucial role in image processing. It is one of the most important features of image. The edge of image is often considered as its fundamental feature in image processing. It has been used in image processing and analyzing technologies such as feature description, image recognition, image segmentation, image enhancement and image data compression on a higher level.In the thesis, we introduce some traditional detection algorithms. What's more, some new image detection algorithms are put off to improve conventional methods' speed and robustness to noise. Its main content is as follows:Firstly, this paper explains some applications of edge detection on practice, and then shows some traditional edge detection algorithms.Secondly, a new algorithm of edge detection base on grey relation is put forward thinking of the edge pixels feature. Using the regulars of a pixel and its neighbor eight pixels, we can easy judge which pixel is the edge pixel.Thirdly, a new algorithm to detect image edge based on local entropy is introduced in this paper. The image edge is detected by local entropy with each pixel and its eight neighboring area pixels. Something is showed by simulation that this algorithm has a height precision in image detection. A better antinoise ability and improving image detection effect are also held by this algorithm because local entropy is produced by all pixels in local window.Finally, a new algorithm to detect image edge based on fuzzy entropy is introduced in this paper. the three natural characters of image edge are that the gray intensity distribution in neighborhood of image is ordered, directive, and structurized. The method is more sensitive for weak edge detection and has better anti-noise performance.