Development Of Image Testing System And Study On Image Algorithm Based On Fuzzy Logic

Image testing, a branch of generalized computer vision, is a practical application of computer vision technology to detection. Compared with traditional testing method, image testing possesses many unique advantages. Just like computer vision in narrow sense, image testing can be divided into two phases, lower-level processing and higher-level processing. As well known, the former is the basis of the latter. Research on image testing emphasizes the analysis and detection of the individual objects in a scene, and the information describing the shape of the objects should be extracted exactly, in order to analyze and measure the objects accurately. Thereby, higher requirements on lower-level processing are needed to image testing.After studying and analyzing special image-detection system and combining the former achievements of our office, the following improvements and new algorithm are suggested, which are based on the characteristics of general image-detection.(1) A general image detection software system is suggested, which is based on structure method. And a new component image processing method is suggested. This system is an open structure software system. New modules can be added to it. Various special image-detection systems can be made up to meet the needs of subscribers.(2) An improved fuzzy logic adaptive enhancement algorithm based on directional information measure is proposed.(3) An image adaptive edge detection method based on fuzzy logic is presented.(4) A new color image-processing algorithm based on fuzzy logic is described, according to the fact that image color information can be fully employed if vector-analyzing method is used in color image processing and analyzing.