Abstract:

For the problem that the lab reports input is timeconsuming and high error rate, an automatic marking system is developed based on character recognition so as to improve the efficiency and accuracy of data input. According to the principle of image recognition, characteristic values are obtained by a series of image preprocessing. The results can be identified by the BP neural network system according to the improved BP algorithm. Access database is used to save the results. After testing, the system can support inputting and saving the lab reports quickly.

Key words: pattern recognition;image recognition;character recognition system;image processing;BP neural network