基于奇异值分解和多尺度小波变换的车牌字符识别研究

Study on Recognizing Characters in Vehicle License-plates based on Singular Value Decomposition and Multi-scale Wavelet Transform

小波变换作为一种新兴的技术,在提取字符的结构特征方面有其自身的优势,而奇异值分解又是一种有效的代数特征提取方法。采用了两级识别方法,粗分类利用多尺度小波变换、特征值分解和3近邻法得到动态候选集,根据该候选集进行细分类。实验结果表明,该方法能够较好地对车牌字符进行分类。

Wavelet transform is a new technology in recent years,and has many advantages in extracting structure features of characters.And algebra features of characters can be effectively extracted by singular value decomposition.In this paper,all characters are analyzed through a coarse classification method and a fine classification method.The experiment result shows that this method can realizes licenseplate characters recognition.