Image Edge Detection Based On Wavelet And Morphology

Edge Detection is one of the basal contents on image processing and analysis, on Which people have not as yet find a satisfactory way out. The edge which is one of the most important fundamental feature of an image, embodying the other features such as position and outline etc., is widely used in image analysis and processing such as feature description, image segmentation, image enhancement, pattern recognition and image compression etc., So Edge Detection is the research hotspot in the technology of image processing and analysis all the while, on which the new theory and methods are put forward constantly.Now commonly used edge detection is mostly using local image differential technology to extract image edge, for example in the physiology and psychology has a very good effect of LOG operator. But using differential technology to extract the edge of the method often noise are very sensitive to noise image edge extraction containing the effect will be discounted. Wavelet edge detection method is in recent years, emerging method, because wavelet transform has good time-frequency localization characteristics and multiscale analysis ability, in the small scales can detect the dispersion-decreasing edge namely Laplace domain shape and roof edges, on larger scales can detect the dispersion-decreasing edge, but also has a natural antinoise characteristics is not sensitive to noise.In order to improve the effect of the edge detection, this paper uses a kind of wavelet analysis and mathematical morphology method to extract the new combination of image edge. The basic idea of mathematical morphology is to use certain shape and structure elements to measure and extracting images of image and corresponding shape, which makes analysis and recognition mathematical morphological edge detection method can effectively remove the noise and protect details, but also has a operation simple, may parallel computing and easy to hardware implementation, etc. This paper based on the correct image based on wavelet decomposition, the high frequency signal enhancement and use of mathematical morphology edge detection method for image edge extraction. This method can effectively underlines the advantages of wavelet and morphology, and to a certain extent improving two methods shortcomings. The main research of this paper as followings:(1)The basic theory of wavelet analysis and digital image edge detection method of classic is discussed in this paper, and given the seven classic method experimental results and analysis.(2)This article through a large number of experimental, in view of the different wavelet base and different decomposition of image enhancement scales, and choose the optimum wavelet and decomposition layers on the high frequency enhance image.(3) Since most of the mixed with a noise in images, image edge and noise with belong to the image of high-frequency information. The paper based on the available morphology of edge detection method of research, puts forward a new multiple structure element anti-noise morphological edge detection method. And classic edge detection method, this paper compares obtained good results. This method for added a noise vehicle images on edge detection, the effect analysis shows that this method in the image edge detection are very good precision and accuracy.