Image Processing Based On Gray System Theory And Mathematical Morphotogy

Image is one of the most important information used by humanity. Image processing refers to using the computer image analysis technology to achieve the required results, or is called imaging, whose main content includes image compression, enhancement and recovery, as well as matching, description and identification. Along with the development of science and technology and mathematics of the times, mathematical image processing technology attracts great attention and gets considerable progress, and has been widely applied in all walks of life all of which has made tremendous social and economic benefitsImage Edge is the most fundamental characteristics of image, which contains the valuable target boundary edge information. These messages can be used for image filtering, image analysis and object recognition. The so-called edge (or fringe) is a collection of the pixels that have a step changes, or changes in the roof of the pool. Edge detection is an important part of the image. Image processing and edge detection algorithm is one of the classic technical problems in the image manipulation. The solution has a very important practical value for us to hold high-level profiling to identify and understand. So, the real edge of as far as possible to remove noise and acquire the image has become a hot topic of research in recent years.This paper carries on discussing from three goals of the image pre-processing, edge detection, and image perspective exposition.First, I pass on some knowledge of the image enhancement in the image preprocessing, especially image enhancement of airspace. It includes the histogram based on the gray-scale image processing points, and Wiener filtering and SUSAN filter of the airspace filtering. Among them, the histogram is based on probability theory, and methods commonly used are histogram and histogram equalizationprovisions, which can increase the dynamic range histogram by amending the gray values, so that we can achieve the overall enhanced image contrast. But the spatial filter template is completed in space images neighborhood operation. According to the characteristics it can be divided into linear filtering, and nonlinear filtering. This paper focuses on the Wiener filtering and SUSAN filtering, and proves that SUSAN filter is better than the Wiener filter on filtering effect by comparing experimental results. Because it can be a very good filtering to preserve the structural features of other objects while filter image noise and satisfy the requirements of smoothing the image noise.Second, edge detection is the most important in the image processing. Detection is directly related to the quality of the image-processing work whether proceeding smoothly or not. For decades, it has been dedicated to the study and solved the problem of how to construct some edge detection operators of good nature and good results. Several traditional edge detection operators, such as Roberts operator, Sobel operator and Prewitt operator which have been developed to deal with structural neighborhood as a center pixel gray-scale image analysis to achieve image Edge Extraction and a better effect. However, these methods also have some shortcomings, such as edge-pixel width, edge positioning accuracy, and more serious noise. Even by the means of de-noising filter, it still will make corresponding fuzzy edge of the side effects difficult to overcome. In addition, because of the physical and illumination,it is on the verge of making the usual images at different scales, so the use of traditional single-scale edge detection algorithm operator is impossible to properly detect the edge of all. There were different types of image itself on the brink, which usually can be divided into non-Slowly and slowly edge. As result of these messages unknown, we called it gray information. This paper does the simulation experiments using MATLAB software on the basis of introducing the principles of classical operator, and withthe results, describes the advantages and disadvantages of several classic operators. Gray System Theory draws the classic characteristics: it has resolved a problem that the introduction of a gradient algorithm computation results in the highly sensitive to noise, and noise often mistakenly seized at risk and marginal quality that detection made is lowered. After Gray theory analysis, a degree of gray incidence in image processing applications is put forward. What's more, Software programming and experimental results show that the feasibility of gray in the image processing , that it has achieved good results in the image edge detection, and that in the gray image processing system it proves the theoretical value and practical value.Third, mathematical morphology was introduced in the description of the image target. Mathematical morphology analyses the image on the basis of geometry whose basic idea is to detect and image feature extraction by using a structural element as basic tools to see the structural elements whether can be placed in an efficient and appropriate internal image. Chapter 4 introduces four basic binary morphological operations. Then Gray morphological transition to the four basic operations of gray scale morphology and put up the expansion to fill method of multi-level test based on the morphology of regional. We describe the target image- using algorithm. Through seeing results it proves the true effectiveness of the algorithm to achieve the desired results.The successful integration of Gray system theory and mathematical morphology will make certain contributions to image processing.