针对古籍汉字结构复杂、风格多变以及字形图像降质所导致传统文字图像检索技术在应用于古籍汉字图像时性能不佳的问题,引入犹豫模糊集理论,提出了一种古籍汉字图像检索模型.首先,设计面向古籍汉字图像的重叠模糊规范化双弹性网格划分,通过考察当前网格与其近邻网格间各种字形要素间的几何和统计特征,定义相应的犹豫模糊元素,进而构成古籍汉字查询图像和目标图像的犹豫模糊集合;其次,以犹豫模糊集合的加权距离测度作为古籍汉字查询图像和目标图像的相似性测度,得到古籍汉字图像检索结果的有序输出.本文算法在11 574幅古籍汉字图像上的检索查准率和查全率分别为78.9%和76.5%.
The traditional character image retrieval technology was difficult to meet the needs of users when it was applied to retrieve ancient Chinese character images because of the complex structure of ancient Chinese characters, changing of font styles, and degradation of font images. To address this problem, a multi-attribute fuzzy retrieval model of Chinese characters in ancient books was proposed by introducing the theory of hesitant fuzzy set. Firstly, an overlapping fuzzy normalized Bi elastic mesh generation for Chinese character images of ancient books was designed. A hesitant fuzzy set containing the query images and target images was established by investigating the geometric and statistical characteristics of various glyph elements between the current grid and its adjacent grids, and defining corresponding hesitant fuzzy elements. Secondly, the weighted distance measure of hesitant fuzzy sets was used to the similarity measure between the query image and the target image, then obtaining the orderly output of the retrieval results. The precision rate and recall rate of retrieval experiments on 11 574 ancient Chinese character images are 78.9% and 76.5%, respectively.
[1] 张睿,丁晓青,方驰.脱机手写汉字识别的最优采样特征新方法[J].中国图像图形学报:A辑,2002,7(2):74-78.DOI: 10.3969/j.issn.1006-8961.2002.02.016.
[2] 姜文,刘立康.基于联合两种特征的手写体维文字符识别[J].计算机工程与应用,2017,53(5):192-196.DOI: 10.3778/j.issn.1002-8331.1507-0207.
[3] 冉耕,黄山,何志辉,等.重叠模糊规范化双弹性网格汉字特征提取[J].计算机工程与设计,2016,37(1):211-215.DOI: 10.16208/j.issn1000-7024.2016.01.040.
[4] 毛晓波,程志远,周晓东.基于特征图叠加的脱机手写体汉字识别[J].郑州大学学报(理学版),2018,50(3):78-82.DOI: 10.13705/j.issn.1671-6841.2017314.
[5] 刘虹,王烈.结合余弦相关性的卷积网络识别汉字的方法[J/OL].计算机工程与应用,2020,56(8):130-135.(2019-04-10)[2020-01-11]. http://kns.cnki.net/kcms/detail/11.2127.tp. 20190408. 1728.006. html.
[6] 郭利敏,葛亮,刘悦如.卷积神经网络在古籍汉字识别中的应用实践[J].图书馆论坛,2019,39(10):142-148.
[7] ZHOU Z Y, ZHAN E Q, ZHENG J B. Stroke extraction of handwritten Chinese character based on ambiguous zone information[C] //2017 2nd International Conference on Multimedia and Image Processing(ICMIP), IEEE, 2017: 68-72. DOI: 10.1109/icmip.2017.45.
[8] 魏玮,郭向丹.基于手写体汉字双弹性网格模糊特征的研究[J].控制工程,2012,19(6):1025-1028. DOI: 10.3969/j.issn.1671-7848.2012.06.023.
[9] MAPARI S, CHAUDHARY N, NAIK S, et al. Usage of fuzzy rule and SOM based model to identify a handwritten chemical symbol or structures[C] //2017 Second International Conference on Electrical, Computer and Communication Technologies(ICECCT), Coimbatore, IEEE, 2017: 68-72. DOI: 10.1109/icecct.2017.8117842.
[10] 柴彦立.基于犹豫模糊特征的古籍汉字图像检索[D].保定:河北大学,2019.
[11] TORRA V. Hesitant fuzzy sets[J]. International Journal of Intelligent Systems, 2010, 25(6):529-539. DOI: 10.1002/int.20418.
[12] 金连文,高学.几种手写体汉字网格方向特征提取法的比较研究[J].计算机应用研究,2004,21(11):38-40. DOI: 10.3969/j.issn.1001-3695.2004.11.013.
[13] ZADEH L A. Fuzzy sets[J]. Information and Control, 1965, 8(3): 338-353. DOI: 10.1016/s0019-9958(65)90241-x.
[14] XU Z S, XIA M M. Distance and similarity measures for hesitant fuzzy sets[J]. Information Sciences, 2011, 181(11): 2128-2138.DOI: 10.1016/j.ins.2011.01.028.
