一种基于参照模型的联机手写汉字识别方法  

Model-based Stroke Extraction and Recognition for On-line Handwritten Chinese Character

作　　者：罗勇[1] 周必水[1] 

机构地区：[1]杭州电子工业学院软件分院,浙江杭州310037

出　　处：《杭州电子工业学院学报》2003年第3期66-69,共4页Journal of Hangzhou Institute of Electronic Engineering

摘　　要：联机手写汉字的识别是模式识别的热点问题,现在所采用的多为特征匹配和结构匹配两种方法。结构匹配方法充分考虑了笔划间的连接关系,能构得到较好的识别率。结构匹配的关键在于如何有效地从输入手写汉字中提取笔划。考虑到同一个汉字的手写体会由于笔划的长短、延伸角度的不同而有所区别,但彼此仍存在极大的相似性,文章提出了一种基于参照模型的手写汉字笔划提取及匹配方法,以解决手写汉字识别问题中的难题。Recognition for On-line handwritten chinese character is a hot issue of pattern recognition, the popular methods of this kind of recognition are feature matching and structure matching. Structure matching method is considered not only the strokes extraction and matching, but also the relation type of those strokes, and this method can get good result in recognition in handwritten chinese character. This paper proposes a model-based stroke extraction and recognition for on-line handwritten chinese character to deal with the difficult problem of how to recognition handwritten chinese character. 

关 键 词：参照模型 联机手写汉字 模式识别 笔划提取 结构匹配 特征匹配 

分 类 号：TP391.12[自动化与计算机技术—计算机应用技术]