复杂背景下基于车牌特征的车牌粗定位实现  

Realization of the license plate location based on the characters in Complex background

作　　者：张鑫[1] 袁得成[1] 郭仁春[1] 

机构地区：[1]沈阳化工学院信息工程学院,辽宁沈阳110142

出　　处：《微计算机信息》2006年第11X期306-308,311,共4页Control & Automation

基　　金：辽宁省教育厅高等学校科学技术研究项目编号:2006A031

摘　　要：本文基于车牌灰度特征和几何特征实现了车牌识别系统(LPRS)中车牌的粗定位。采用直方图Ostu方法选取阈值进行二值化、采用3*3大小的十字形窗口进行中值滤波,分别用水平投影和垂直投影法寻找车牌的左右边界和上下边界,结合图像灰度序列与单位矩形序列的互相关函数定出车牌的位置,为以后车牌进一步精确定位以及字符识别打下基础。This article realized the license localization in the car license recognition system （LPRS）based on the car license gradation characteristic and the geometry characteristic. Using the histogram Ostu method to select a threshold in binarization, using the 3＊3 size-cross-window value filter, using horizontal project and vertical project to seek the car license boundary, then through the mutual correlation function of union picture gradation sequence and the unit rectangle sequence decide on the car license the position, building the foundation for later car license further precise localization and the character recognition.

关 键 词：车牌定位 二值化 投影法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]