Study And Implementation Of Neural Network Image Recognition

The technology of Neural Network image recognition is a kind of new-style image recognition technology, which is developed due to the theory of present computer technology, image processing, artificial intelligence and pattern recognition. This article concisely introduces the theory of image recognition and some kinds of Neural Network models commonly used in pattern recognition; it sets emphasis on the introduction of feed-forward Neural Network model and BP algorithm. Both the advantages and disadvantages of the BP algorithm are discussed. According to the disadvantages, some improved BP algorithms are introduced. The method of BP network classifier design, the way of sample selection and organization are both delivered. The Neural Network image recognition without feature extraction overcomes the difficulties of feature extraction under confused backgrounds. The result of the simulation experiments indicates that image recognition method based on neural network without feature extraction is feasible. It proposes a dynamically reconfigurable FPGA realization model of the BP network. Although such system has not yet been applied to the practical image recognition, it paves the way for the following research.