基于自适应小波神经网络的车牌融合识别算法

A License Plate Fusion Recognition Algorithm Based On Wavelet Neural Networks

该文提出了一种基于自适应小波神经网络的车牌融合识别算法。该算法首先根据车牌中字符的形态特征和横向纹理属性,通过小波分析来提取字符的小波统计特征,对自适应小波神经网络进行训练并识别车牌字符;然后,对单神经网络识别结果提取特征值并参与决策层融合,最终获得识别结果。实验结果表明,该文提出的算法是一种切实可行、准确高效的方法,对复杂背景下拍摄的汽车牌照的字符识别具有很好的鲁棒性、容错性。

In this paper,a novel fusion algorithm is proposed for vehicle license recognition which is based on adaptive wavelet neural networks.Firstly applying wavelet transform to preprocess vehicle license characters as textural features,character statistic features are extracted with wavelet analysis of license characters,then training the neural networks.Lastly a neural network is used to fusion information.Experimental results demonstrate that the proposed approach could efficiently be used as a vehicle license characters fusion recognition system with high convergence,which is robust for background complexity.