Character recognition algorithm bas

Character recognition algorithm based on neural network generally includes four steps: preprocessing, feature extraction, classifier training and recognition. The key step is to train the neural network classifier with the extracted features after extracting the features of the characters to be recognized. This neural network method makes full use of the classification ability of neural network. At the same time, it faces two difficulties: too much feature parameter extraction will increase the training time of classifier, too little will lead to the reduction of recognition rate. Another method is that it does not need feature extraction, and makes full use of the characteristics of neural network. The feature extraction is realized automatically by neural network. It only needs to input the image to be processed into the network directly. This method has good anti-interference performance and high recognition rate. However, the disadvantage of this method is that the increase of input pattern dimension may lead to the network scale and dimension disaster, and the amount of information to be processed is too large. Although the recognition rate is high in recognition effect, it also has strong fault-tolerant ability, and the classifier can be continuously improved to improve the recognition rate, but because of its slow recognition speed, it is difficult to meet the requirements of real-time in practice Use.