Research On Vehicle Plate Recognition Technology Based On Neural Network

Vehicle plate recognition technology is a key technology of the highway automatic charge system,city road supervisory system,the intelligent parking lot management system and public security of law enforcement system.Vehicle Plate recognition technology can be used in the field such as road charge and traffic.It's application will be helpful to save manpower cost,raise efficiency and improvement to manage system.Along with the quick increase of the automobile quantity of our country,vehicle plate recognition technology presents huge economic value and realistic meaning.Based on the prior work, LPR system software is improved, we made further research focusing on theories and technologies of license plate characters segmentation and recognition, character feature extraction and recognition is our concern.The Principal Component Analysis method is applied in the feature extraction of license plate charaeters,the principal component is reduced by Rough Set,the fewer features are taken as input of neural network,thus not only the complexity of network strueture is decreased,but the rate of learning and the recognition rate of characters are enhanced.In the research of recognition methods, the advantages and disadvantages of model match and neural network are compared and analyzed.We combined these two methods and designed some classifiers using a multilevel-multiclassification scheme, which can classify numbers, letters, alphanumeric charaeters and Chinese characters.The result shows that it is more effectively in similar character recognition.