基于阈值分割和区域生长的车牌识别方法*
Vehicle License Plate Recognition Method Based on Threshold Segmentation and Region Growing

 

胡伟平, 王日凤
HU Weiping, WANG Rifeng
 

(广西科技大学计算机科学与通信工程学院,广西柳州   545006)
(School of Computer Science and Communication Engineering,Guangxi University of Science and Technology,Liuzhou,Guangxi,545006,China)

 

摘要:【目的】改善车牌定位的质量,提高车牌识别的正确率和效率。【方法】联合使用阈值分割和区域生长算法进行车牌定位,使用垂直投影法进行字符分割,并使用字符模板匹配方法实现车牌字符的识别。【结果】基于阈值分割与区域生长的车牌识别方法能准确地识别出车牌号,识别率高,运行速度快。【结论】该方法实时性较好,具有一定的实用价值。
关键词:阈值分割   区域生长   垂直投影   车牌识别
中图分类号:TP181      文献标识码:A      文章编号:1002-7378(2016)01-0054-05

Abstract:【Objective】To improve the effect of vehicle license plate location algorithm, and the accuracy and efficiency of vehicle license plate recognition algorithm.【Methods】Threshold segmentation and region growing algorithm were combined to locate the vehicle license plate area. Vertical projection algorithm was used to split the plate characters. Template matching algorithm was used to recognize the plate characters.【Results】Experimental results show that the accuracy and speed of license plate numbers recognition are both improved through our method.【Conclusion】The accuracy and realtime of vehicle license plate recognition method based on threshold segmentation and region growing shows its practical value.
Key words:threshold segmentation,region growing,vertical projection,vehicle license plate recognition

 

*广西科技大学科学基金项目(校科自1419212),国家自然科学基金项目(61440017),广西高校一般资助项目(201203YB129)和广西多源信息挖掘与安全重点实验室开放基金项目(MIMS13-04)资助。

 

PDF全文下载cbimg.cnki.net/Editor/2016/0214/gxkx/cd558230-f120-440a-8168-55a2ec258904.pdf

发布日期:2016/2/14