Car plate number characters recognition using Gabor orientation features and neural networks

A new plate number character recognition method based on Gabor orientation feature and neural networks technologies is presented in this paper. Based on elastic meshing techniques and Gabor filter, a new feature extraction approach for low resolution plate number characters is proposed. An integrated neural networks model is designed as an intelligent classifier. Experiments on large data set (1000 sets of training data and 500 sets of testing data) shows the recognition rate of 99.26%, suggesting that the proposed approach is very effective.