首页 | 本学科首页   官方微博 | 高级检索  
     

复杂背景下彩色图像车牌提取与字符分割技术
引用本文:陈锻生,谢志鹏,刘政凯.复杂背景下彩色图像车牌提取与字符分割技术[J].小型微型计算机系统,2002,23(9):1144-1148.
作者姓名:陈锻生  谢志鹏  刘政凯
作者单位:1. 中国科学技术大学,电子工程与信息科学系,安徽,合肥,230027;国立华侨大学,计算机科学系,福建,泉州,362011
2. 国立华侨大学,计算机科学系,福建,泉州,362011
3. 中国科学技术大学,电子工程与信息科学系,安徽,合肥,230027
摘    要:车牌图像分割是车辆牌照自动识别系统的关键部分,本文提出和提取了车牌标准模式的多种重要特征,综合利用数学形态学运算、连通域形状测量和彩色图像分析,有效地解决了复杂背景下提取车牌的问题,并实现了牌照的倾斜校正和边框线与铆钉去除,最后分离出牌照的各个字符、实验表明该技术方案在实时性,鲁棒性和正确率都达到实用技术指标。

关 键 词:复杂背景  彩色图像  车牌提取  字符分割  图像分割  连通域测量  数字图像处理  计算机
文章编号:1000-1220(2002)09-1144-05

Extraction of Number Plate and Character Segmentation From Color Image Under Complex Background
CHEN Duan sheng ,XIE Zhi peng ,LIU Zheng kai.Extraction of Number Plate and Character Segmentation From Color Image Under Complex Background[J].Mini-micro Systems,2002,23(9):1144-1148.
Authors:CHEN Duan sheng    XIE Zhi peng  LIU Zheng kai
Affiliation:CHEN Duan sheng 1,2,XIE Zhi peng 2,LIU Zheng kai 1 1
Abstract:Number plate extraction is the key component of automatic vehicle license recognition system. Several important features of standard number plate are designed and extracted in this paper, and the problem of extracting number plate under complex background is resolved using the mathematic morphological operation, shape measurement of connected area and color image analysis, thereafter putting right the slant plate and getting rid of the outer bounding frame and rivet, finally extract the separate car license character. The experiment result shows that the proposed approach is good enough to be put into practical use, with respect to real time, robustness and accuracy.
Keywords:measurement of connected area  color image analysis  extraction of number plate
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号