Research On Image Edge Detection Methods

Digital image edge detection plays an import part in image analysis, such as image segmentation, interested region recognition and region shape extraction. And it's an import method in image feature extraction of image recognition. The edge includes the valuable information of the image which can be use in image understanding and analysis. And through edge detection, we can greatly reduce the calculation of image analysis and processing in the following step. Usually, the first step of image understanding and analysis is edge detection, and it has been the most active topic in the machine vision research field, also it plays an import part in engineering application.Most of the traditional edge detection algorithms, such as Roberts, Sobel, Prewitt, Kirsch, Laplace, just construct an edge detection algorithm with a small neighborhood in each pixel of the original image, and then carry out with first differential or second differential operator in order to obtain the maximum gradient or the zero-crossing point of the second derivative, finally select an appropriate threshold to extract the edge. But these algorithms share the same shortcomings, for example, they are sensitive to noise, they can't select threshold adaptively, and the detection results are not so well.In this paper, we do a deep research on the edge detection theory and algorithm, base on analyzing the traditional edge detection algorithm in detail, we focus on Canny algorithm, combined with MTM algorithm and Otsu algorithm to improve the filtering method and dual threshold selection method in Canny algorithm, and we call the improved algorithm CMO for short. Finally, we use VC++6.0 to implement CMO algorithm, and the experiment results show that CMO algorithm can get better results than traditional Canny algorithm.