计算机技术与发展Issue(5)：195-198,4.DOI:10.3969/j.issn.1673-629X.2013.05.050
基于最小二乘支持向量机车牌字符特征识别
LSSVM-based License Plate Character Feature Recognition
摘要
Abstract
LP recognition is analyzed and researched in LP image preprocessing,location and character segmentation,feature extraction and feature classification. Among them,feature extraction and classification is the key link to reach more better recognition rate and speed. Least square support vector machine (LSSVM) is a kind of novel machine learning method. Propose a new method of license plate (LP) character recognition based on LSSVM and singular value decomposition (SVD). This method is based on preprocessing,extracts the singular value features of the LP character after segmentation,the main feature is compressed and contained,uses LSSVM to classify and identify. Experiment is based on LP image. The experimental results demonstrate the efficiency of the proposed approach.关键词
车牌识别/最小二乘支持向量机/奇异值分解Key words
license plate recognition/least square support vector machine (LSSVM)/singular value decomposition (SVD)分类
信息技术与安全科学引用本文复制引用
刘静..基于最小二乘支持向量机车牌字符特征识别[J].计算机技术与发展,2013,(5):195-198,4.基金项目
国家统计局课题项目(2012LY056) （2012LY056）
2011年度军民融合研究基金项目(11JMR09) （11JMR09）
2012年陕西省科技计划项目(2012JM8031) （2012JM8031）
陕西省自然科学基金项目(2011JM8020) （2011JM8020）
渭南师范学院科研计划项目(12YKZ053) （12YKZ053）
