Research Of Edge Detection For Grey-Scale Image Based On Mathematical Morphology

Mathematical morphology founded on set theory is a new method applied in the field of image processing and pattern recognition. Its prime principle is using a certain structuring element to measure and extract the corresponding form in an image so that we can analyze and recognize the image. Edge detection plays a special role in image processing and computer vision. It is one of the most important links in the basic vision processing. The edge detection processing based on mathematical morphology excels that based on differential operation. It is not so sensitive to noise as differential operation and the extracted edge is smoother. It can not only embody the set characteristics of an image and efficiently detect the edge but also meet the real-time expectations. At the same time, it can reduce the noise by changing morphological scale.Above all in this paper the basic theory of mathematical morphology is discussed, its application in image processing is analysed and introduced as well. Then the basic morphological filter, the generalized morphological filter and the morphological filter with omnidirectional structuring element are expounded in detail. Considering the advantages of mathematical morphology exploited in image edge detection, a multi-scale morphological edge detection algorithm is proposed in this paper. At the end a new antinoise edge detector is presented by improving the edge detector mentioned above in order to deblur the detected edge efficiently and eliminate the influence of noise appropriately.The simulated result demonstrates that the edge detection algorithm has a good performance of noise reduction and real-time compared with traditional edge detectors, and the detected edge is smooth. So this method has its practicability.