摘要
车牌字符识别是车牌智能识别系统中的关键技术之一,为了提高车牌字符识别准确率和实时性,提出了一种基于车牌字符笔画斜率特征的字符识别算法,新算法在对准确分割后的车牌字符图像进行归一化、二值化和细化等预处理后,提取车牌字符笔画像素点的斜率值,再将斜率值构造成识别向量,进而通过计算特征向量的夹角值实现待识别字符与标准模板字符的相似度判断;实验结果表明,新的算法与已有的传统字符识别算法相比,既提高了识别准确率,又降低了识别耗时,结果满足实际工程应用需要.
License plate character recognition is one of key technologies in intelligent recognition system of license plate, in order to improve the accuracy and real-time performance of character recognition of the license plate, a kind of character recognition algorithm is proposed based on slope feature of the strokes of Chinese characters.After the processing of normalization, binarization and thinning and so on of accurately segmented character images of license plate, the new algorithm extracts the slope value of the pixel points of the strokes of Chinese character of license plate, then slope value is composited into recognition vector, and furthermore, the similarity judgment between recognition character and standard template character is realized by calculating the angle value of characteristic vector. Experiment results show that the new algorithm, comparing with the existed traditional character recognition algorithms, not only improves recognition accuracy but also decreases recognition time consumption, so as to meet the demand of practical engineering application.
出处
《重庆工商大学学报（自然科学版）》
2014年第9期72-76,共5页
Journal of Chongqing Technology and Business University:Natural Science Edition
基金
黄山学院校级科学研究项目(2012xkj004)
关键词
字符笔画
斜率
向量
字符识别
strokes of Chinese character
slope
vector
character recognition
