基于Fisher准则特征融合策略的PNN车牌汉字识别_高全华.pdfVIP

: Fisher PNN 基于Fisher 准则特征融合策略的PNN 车牌汉字识别 1 1 2 高全华 , 张士勇 , 孙锋利 (1. , 　710064;2. , 　710077) 　:为克服图像二值化预处理的不利影响, 提高汉字字符图像特征的表征 力, 降低传统算法的训练复杂度, 在此 直接应用车牌字符灰度图像, 基于Fisher 准则对提取的Pseudo-Zernike 矩特征和Gabor 变换特征进行融合, 在最佳鉴别矢量 方向上求得表达 力更强的融合特征。 训练结构简单、速度较快的概率神经网络作为识别器, 实现车牌汉字的正确、快速识 别。 试验表明, 此算法无需对车牌汉字图像二值化, 与所采用的概率神经网络识别器相比, 传统BP, SVM 等的训练复杂度降 低, 速度提高, 经表征 力更强的融合特征训练后, 具有更高的识别准确率。 :Fisher 准则;鉴别矢量;特征融合;概率神经网络;车牌汉字识别 :TP391　　　　　:A 　　　　　:1004-373X(2010)0 -0106-05 PNN Recognition of Chinese Characters on License Plate Based on Fisher Discriminant Criterion Feature Fusion Strategy 1 1 2 GAO Quan-hua , ZHANG Shi-yong , SUN Feng-li (1.College of Science, Changan U niversity, Xian 710064, China; 2.college of Electronic and Information, Northwest Polytechnic University, Xi an 710077, China) Abstract:A novel algorithm for recognizing Chinese characters on license plate is proposed so as to achieve three purpo- ses:overcoming disadvantages of two-value preprocessing, raising the characterization ability of image features and decreasing training complexity of traditional algorithm s.According to Fisher discriminant criterion, tw o kinds of image features, Pseudo- zernike moments and Gabor transforming coefficients are fused along the direction of the optimal discriminant vectors.The new fused feature has better performance to characterize the image.Probabilistic neural network(PNN)is trained by new fu- sion features as the classifier due to its simple structure and quick learning rate.Numeral experiment show s this algorithm does not need two-value preprocessing and the obtained clas