复杂背景下的字符模板匹配改进算法研究  

Study on Algorithm of Improved Character Template Matching Under Complex Background

作　　者：杨沐[1] 王晨升[1] 陈志强[1] 陈亮[1] 李磊[1] 

机构地区：[1]北京邮电大学自动化学院,北京100876

出　　处：《软件》2012年第12期97-100,共4页Software

摘　　要：本文论述了针对复杂背景铸件字符图片,如何将其中的字符进行准确识别的关键算法。整个文章分为五个部分。首先,在引言部分介绍了项目背景;其次,介绍了整个项目的系统结构及各个模块的职责;再次,详细描述了图像处理及匹配的核心算法,包括字符图像网格匹配算法、图像预处理参数的自适应匹配算法;最后,概括总结。本文论述的核心内容是讲述基于图像预处理部分结合图像匹配部分的参数适应匹配算法以及网格匹配算法。This paper discussed the key algorithm of how to accurately identify the characters in a complex background picture. The whole article is divided into five parts. Firstly, It describes the project background in the introductory section; Secondly, It introduce the entire system structure and the responsibilities of the various modules; Then, It detailed describes the core algorithm of the image preprocessing and matching; Finally, It come to the summary. The core content of this article is about Grid matching algorithm and the parameter adaptation matching algorithm based on the image preprocessing part and the image matching part.

关 键 词：铸件打码 模板匹配 特征化 适应匹配 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]