基于区域生长算法的汉字笔画统计与分析  

Stroke Statistics and Analysis of Chinese Characters Based on Region Growing Algorithm

作　　者：蔡志伟 奚海丹 田云松 CAI Zhi-wei;XI Hai-dan;TIAN Yun-song(School of Computer Science and Engineering,Dalian Minzu University,Dalian Liaoning 116605,China;Dalian Chinese Font Design Technology Innovation Centre,Dalian Minzu University,Dalian Liaoning 116605,China;Shenyang Open University,Shenyang Liaoning 110003,China)

机构地区：[1]大连民族大学计算机科学与工程学院,辽宁大连116605 [2]大连民族大学大连市汉字计算机字库设计技术创新中心,辽宁大连116605 [3]沈阳开放大学,辽宁沈阳110003

出　　处：《大连民族大学学报》2023年第3期261-264,288,共5页Journal of Dalian Minzu University

基　　金：辽宁省自然科学基金项目(2020-MZLH-19);贵州省科技支撑计划项目(2021-534)。

摘　　要：针对现阶段汉字笔画数据集划分时所含笔画类别较少的问题,设计符合汉字特征的生长控制策略和算法框架,实现汉字笔画小类别的划分。同时,采用统计学方法,对汉字不同笔画的出现频率进行分析,并探究笔画与汉字语义之间的联系,构建了新的汉字笔画数据集,为字体设计中汉字笔画拼接奠定基础。使用GB2312编码和Unicode编码对《信息交换用汉字编码字符集》中的6763个汉字及32类笔画进行编码。通过对样本数据的实验验证,算法在汉字笔画的识别和统计分析方面表现良好,构建的汉字笔画数据集为汉字的研究以及文化传承提供了有力的技术支持。In view of the problem that there are a few stroke categories in the division of Chinese stroke data set at the present stage,a growth control strategy and algorithm framework with Chinese character characteristics should be designed to realize the division of small categories of Chinese strokes.At the same time,statistical methods are used to analyze the occurrence frequency of different strokes of Chinese characters,the relationship between strokes and Chinese semantics is explored,and a new data set of strokes of Chinese characters is built,which lays a foundation for the stitching of strokes of Chinese characters in font design.GB2312 encoding and Unicode encoding are used to encode 6763 Chinese characters and 32 strokes in the Coded Character Set of Chinese Characters for Information Exchange.Through the experimental verification of sample data,the algorithm performs well in the recognition and statistical analysis of Chinese stroke,and the constructed Chinese stroke data set provides strong technical support for the research of Chinese characters and cultural inheritance.

关 键 词：汉字笔画 区域生长算法 数据集 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]