[1]陈泓汉,王 涛,熊显航.二阶段手写汉字骨架提取优化[J].计算机技术与发展,2023,33(07):41-46.[doi:10. 3969 / j. issn. 1673-629X. 2023. 07. 006]
　CHEN Hong-han,WANG Tao,XIONG Xian-hang.Two-stage Handwritten Chinese Character Skeleton Extraction Optimization[J].,2023,33(07):41-46.[doi:10. 3969 / j. issn. 1673-629X. 2023. 07. 006]
点击复制

二阶段手写汉字骨架提取优化()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
33
期数:
2023年07期
页码:
41-46
栏目:
媒体计算
出版日期:
2023-07-10

文章信息/Info

Title:
Two-stage Handwritten Chinese Character Skeleton Extraction Optimization
文章编号:
1673-629X(2023)07-0041-06
作者:
陈泓汉1 ; 王 涛2 ; 熊显航2
1. 华南师范大学 人工智能学院,广东 佛山 528200;
2. 华南师范大学 计算机学院,广东 广州 510631
Author(s):
CHEN Hong-han1 ; WANG Tao2 ; XIONG Xian-hang2
1. School of Artificial Intelligence,South China Normal University,Foshan 528200,China;
2. School of Computer Science,South China Normal University,Guangzhou 510631,China
关键词:
手写汉字; 细化算法; 区域畸变; 分叉点合并; 畸变修复
Keywords:
handwritten Chinese characters; thinning algorithm; area distortion; merge branching point; branching area distortion
分类号:
TP301. 6;TN957. 52
DOI:
10. 3969 / j. issn. 1673-629X. 2023. 07. 006
摘要:
骨架提取是研究手写汉字的重要先决步骤,基于细化算法提取手写汉字骨架存在笔画交叉处产生不同程度畸变以及毛刺等问题,提出一种适用于细化算法的二阶段手写汉字骨架提取优化算法。 使用细化算法得到待修复的汉字骨架,通过二阶段检测算法识别汉字骨架中笔画交错分叉区域以及确定畸变待修复范围,并依据分叉区域类型,采用相应的修复策略。 实验结果表明,该二阶段检测法能够有效地检测出笔画分叉区域,准确率达到 96. 89% ,且分叉区域归类算法可很好地识别出畸变区域类型。 经过修复后的汉字骨架能够正确地反映笔画间的原有拓扑结构,使得最终优化骨架更能突出手写汉字特征,验证了算法的有效性。
Abstract:
Skeleton extraction is an important steps for the study of handwritten Chinese characters. There are some problems in extractingthe skeleton of handwritten Chinese characters based on the refinement algorithm,such as distortion and burr at the cross of strokes. Inorder to solve the problem,we propose a two-stage handwritten Chinese character skeleton extraction optimization algorithm. A thinningalgorithm is used to obtain the original skeleton. Afterward,identiying the stroke intersection area in the skeleton and determining the distortion range to be repaired by two - stage detection algorithm. According to the type of fork area,the corresponding repair strategy isadopted. The experimental results show that the two -stage detection method can not only effectively detect the stroke bifurcation areawith the accuracy rate of 96. 89% ,but also well identify the type of distorted region. The optimized skeleton can better highlight thefeatures of handwritten Chinese characters. The repaired skeleton of Chinese characters can accurately reflect the original topologicalstructure between strokes, so that the fina

相似文献/References:

[1]王娜 马昕.基于细化算法的移动机器人拓扑地图创建[J].计算机技术与发展,2009,(10):11.
　WANG Na,MA Xin.Mobile Robot Topological Map Building Based on Thinning Algorithm[J].,2009,(07):11.
[2]罗湘 韦玉科.基于多结构元的数学形态学的边缘检测[J].计算机技术与发展,2012,(04):85.
　LUO Xiang,WEI Yu-ke.Morphological Edge Detection Based on Multi-Structuring Elements[J].,2012,(07):85.
[3]陈 品,王 涛*,张德港.一种基于模板的二阶段 ZS 细化算法[J].计算机技术与发展,2023,33(02):44.[doi:10. 3969 / j. issn. 1673-629X. 2023. 02. 007]
　CHEN Pin,WANG Tao*,ZHANG De-gang.A Two-stage ZS Thinning Algorithm Based on Template[J].,2023,33(07):44.[doi:10. 3969 / j. issn. 1673-629X. 2023. 02. 007]

更新日期/Last Update: 2023-07-10