基于标注点分析字体结构与系统设计  

Analysis of font structure and labeling system design based on annotation points

作　　者：律睿慜 席旭 Lü Ruimin;XI Xu(School of Artificial Intelligence and Computer,Jiangnan University,Wuxi 214122,China)

机构地区：[1]江南大学人工智能与计算机学院,江苏无锡214122

出　　处：《传感器与微系统》2022年第12期90-94,共5页Transducer and Microsystem Technologies

基　　金：江苏省文化科研基金资助项目(19YB48);2020中国江南与大运河文化带建设重点研究项目(JUSRP12089)。

摘　　要：为了更准确地分析书法结构的特点,提出一种基于标注字体关键点分析结构的方法。首先,选取间架不同的汉字进行研究,并以颜、柳、欧、赵等楷书作为研究书体。然后,参考LabelImg等数据标注工具,实现一种能采集字体关键点的标注系统。最后,利用众包标注实验获取的关键点标注值计算分析字体结构。结果表明:利用字体标注关键点的距离与角度进行统计分析能得出其结构特点,设计的系统也可以应用到书法图像的标注制作与结构计算中,而关键点可以作为一种字体结构量化研究的新方法。In order to analyze on characteristics of calligraphy structure more accurately, a method of structure analysis based on key points annotation is proposed.Firstly, choose different forms of Chinese characters and take standard script of Yan, Liu, Ou, Zhao styles as research handwriting.Then, reference data annotation tool, LabelImg, a labeling system is realized.Finally, the font structure is calculated and analyzed by using the annotation value of key points obtained from crowdsourcing annotation experiment.The result shows that structure features are obtained by statistical analysis on distance and angle between labeling key points of font.The proposed system can also be applied in label making and structural calculation, and key points can be used as a new method for quantitative study on the character structure.

关 键 词：字体结构 关键点 标注系统 人文计算 BEZIER曲线 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论] TP391[自动化与计算机技术—计算机科学与技术]