摘要
车牌识别是智能交通系统中的重要组成部分,主要包括车牌定位、车牌字符分割和车牌字符识别3部分。本文使用MATLAB对车牌识别进行了研究,提出了一种基于多特征组合的车牌识别算法。该算法首先通过边缘检测算子和数学形态学运算得到车牌候选区,然后利用车牌的矩形特征、宽高比特征和垂直投影特征对这些候选区域进行筛选,并定位出车牌;二值化、滤波和形态学开运算后使用投影二分法分割出7个车牌字符;最后使用模板匹配和特征统计相结合的方法识别出车牌字符。实验表明该算法是有效的、可行的,能够在复杂背景下准确的定位和识别出车牌字符。
License plate recognition plays an important role in Intelligent Transportation System,mainly including license plate location,license plate character segmentation and license plate character recognition.This paper studied the license plate recognition with MATLAB,and proposed a license plate recognition algorithm based on multi-features.This method firstly confirmed vehicle license candidate regions with edge detection operator and mathematical morphology operations,then screened these candidate regions with license plate rectangular features,aspect ratio features and characteristics of the vertical projection,and located license plate.After binarization,filtering and morphological opening operation,seven license plate characters were segmented by projection dichotomy.Finally,license plate character was recognized by the combining method of template matching and statistical characteristics.Experiment demonstrates that this method is feasible,effective and can accurately locate and identify the license plate characters in the complex background.
出处
《常州大学学报（自然科学版）》
CAS
2010年第4期47-51,共5页
Journal of Changzhou University：Natural Science Edition
基金
江苏省高校自然科学基金(08KJB520002)
关键词
图像预处理
车牌定位
车牌字符分割
车牌字符识别
image preprocessing
license plate location
license plate character segmentation
license plate character recognition
