Color Morphology New Model And Application Research

Color is a reaction phenomenon of eyes to light. Color image is made up of vector data which is constituted by three-channel(or multi-channel) of components, so color image often can provide more information than the gray image. Color image processing is more and more get people’s attention and widely used in computer vision, pattern recognition, multimedia communications and other main fields, therefore, the study of color image processing is of great research value. Mathematical morphology is quite a practical image processing method, and for the processing of color image morphology research has draw more and more attention to many scholars. This paper foucus on color image morphology based on graph and application processing system research.The author’s major contributions are outlined as follows:1.The new graph space color vector mathematical morphology processing model have been studied. In the traditional mathematical morphology, the basic dilation and corrosion operators are mainly processed to find the maximum and minimum in the collection of structure element. The model maps color image to graph space, which makes the vector pixels in the color image as the basic vertices, the relationship between vector pixels as edges, building the graph structure contains vector vertices and edges. At the same time, combined with the new model this paper proposes a new color vector extremum extraction algorithm. Based on the graph structure the algorithm constructs minimum spanning trees, selects nodes which are not cut points iteratively, until the rest is two nodes. And they are the maximum and minimum.2.Designing the new color vector morphological operators. Based on graph space new model and vector extraction algorithm of color image morphology, this paper design four new basic color vector morphology operators, including dilation, errosion, opening and closing. And then they are applied to color natural images and IC defect image processing, the experimental results showed that morphological operator based on the new model can well protect the integrity of the color information, effectively avoid the emergence of a new color component, protect the topological structure of the image.3. Studying color vector filtering and edge detection based on the new model. Theory withpractice is an important step to improve the theoretical results. When processing color image by traditional mathematical morphology method, it is inevitable to meet the problem of processing vector data and lead to the result of filtering and extraction having fuzzy color, poor effect and other problems. Based on the new model, this paper proposes color vector morphological filtering and edge detection operator in use of the comprehensive characteristics of mathematical morphological erosion, dilation, opening and closing operation. The experimental results show that, the operator proposed by this paper can effectively complete filtering and extraction under the situation of guaranteeing color equilibrium distribution, which is remarkably improved compared with the traditional methods.4.Putting forward the color vector morphological low illumination image enhancement processing technology. Due to the difference of detail gray in low illumination image is very small, and the grey value of the image is very low, it is difficult to processing the image. Therefore, this paper combines weber theorem, uses morphological erosion and open operator, proposes color vector morphological operator based on the new model. The experimental results show that, the color vector morphology processing method in this paper can complete the task of image enhancement globally and effectively and process the details locally.