Image Processing Based On Fuzzy Analysis And Its Application In Non-Destructive Detection

Computer image processing technology is an important part of image project. It is applied extensively in aerospace, traffic, communication project, document processing and industrial detection and so on. In recent years, the application of fuzzy analysis has made progress in image processing. Because of the vagueness of the actual image, Application of fuzzy analysis in image processing generally can reach good processing effect.Nondestructive detection applies physical method to carry out detection base on the prerequisite of not damaging the internal structure of the object. The application of computer has speeded development of nondestructive detection technology. The real time images of objects are collected by means of ray, and then the images are processed with the image technology to optimize the image quality.This paper discusses the image processing techniques based on fuzzy analysis and the application problems in nondestructive detection. And in this paper, three aspects of image processing are researched as follows:Firstly, the processing object of the paper is the digital image gained from the X ray . The image has plenty of random noise, so at first the noise should be reduced. Paper has discussed some methods of reducing the noise used frequently, including the operation of using template and overlay processing and so on. Through comparing various processing effect and combines actual object, the method of reducing the noise based on the statistical information of each pixel was put forward. It has referred to the idea of image overlay processing, at the same time, has considered the randomness of noise. The experimentation proofs the effect of this method is good.Secondly, to be convenient for person to identification, the contrast of the image has been enhanced. In strengthen processing chapter, some strengthen methods used frequently have been discussed, including histogram balanced processing and template processing and so on. This chapter has discussed emphatically the vague strengthen processing technique, and has put forward the improvement vague strengthen algorithm. The new algorithm has accelerated processing speed based on not influencing strengthen effect and has gotten satisfactory processing effect.Finally, the paper discusses the image segmentation. Some general image segmentation algorithms have been summarized first, and then the FCM algorithm has been discussed emphatically. In the realization of this algorithm, the author combine the histogram information of the object, put forward some own methods about the initial assemble center and the assemble kind. The algorithm has obtained better effect.