The Research Of Intelligent Identification System Based On X-ray

Judging X-ray film's quality is an important part in X-ray detection. Traditionally, the judging work is done by judging worker, but it's a very large work.It has many disadvantages, for example, strong light will hurt the eyes, low efficiency, and the defection of the technical staff assessed the quality,experience and the impact of external conditions,results often vary from person to person.Artificail-decfection can not meet growing development needs.With the rapid developmentof compupters,digitial image processing technologies,digital film, intelligent detection has become a non-destrctive direction.In the X-ray film to solve the digital issue,the image defection identification and classification of films on a smart assessment become a priority. The paper seeks to build a intelligent identification system,in the image of the knowledge base construction on defects and deficiencies identified classification reasoning done a lot of work on the study.Through much data collection, the knowledge base of knowledge in this paper is divided into two parts, shallow knowledge base and deep knowledge base. Shallow knowledge base is a standard fuzzy BP neural network, knowledge that is implicit within, from the shape of image defects and deficiencies image category, can be obtained through the learning network-related parameters such as the right matrix and vector threshold. Deep knowledge base is reflected in two aspects: First, the self-knowledge, the learning is not passed directly from the experience of the acquisition of knowledge. Defects in the pattern recognition fuzzy on this technology and artificial neural network technology, select fuzzy BP neural network learning algorithm. Through the simulation test and prove that neural networks and fuzzy BP BP network shortcomings of traditional recognition rate as compared to the fuzzy theory training samples so that the mapping network capacity and improve the ability of the network. Therefore, the ability to identify the former than the latter is much higher. When the shallow knowledge base can not be given the correct identification (identification of Uncertainty), the system through the activation of deep knowledge base for reasoning, come to recognize the results. Inference against the information in the database, selected in the knowledge base on the current seam of useful knowledge reasoning judgement.In this paper, designed by digital X-ray film defects intelligent identification system after a large number of simulation and part of the experiment, was more satisfactory results.