X-ray nondestructive test digitized technology is one of hot research fields in recent years. Its development and application enhance the level of industrial automation and intelligent test and have very important practical significance. The theory and application of the digital image processing technology used in the nuclear pressure vessel nondestructive test negative image process are discussed in this paper. When inspectors look up the faults in the negative image in light, taken from the nuclear industry, they may meet shortage of information because of the limitation of eyes and of the high density of the negative gray. Firstly, the whole composed principal and system processing flow of the image checking system are introduced in the paper. Secondly we mainly study the material application of image transforming, image strengthen , image compressing and the letter recognition. According to the character of human being and the negative image, the processed image that are suit to be checked and recognized can be got by the use of different algorithms. How to choose an efficient algorithm to compress the image is one of main research in the paper, because it takes a lot of time and space to save and transmit the image data at a time. The united method of template match and neural network is demonstrated in the paper and by the way the letter character information in the image can be recognized and classified. The correct ratio and the recognized effect are satisfied to the practical need. The level of intelligence and information is lifted up. The processing system is studied and built up with VC++ to process the information and to save and manage the image. The whole system process is presented in the paper, such as software algorithm, system design and the software and hardware environment of the system. At last the shortage is put forward and the way to get better is forecasted.
|