Self-information of radicals: A new clue for zero-shot Chinese character recognition

doi:10.1016/j.patcog.2023.109598

CORC > 自动化研究所 > 中国科学院自动化研究所 > 多模态人工智能系统全国重点实验室

	Self-information of radicals: A new clue for zero-shot Chinese character recognition
	Luo, Guo-Feng 3,4; Wang, Da-Han 3,4; Du, Xia 3,4; Yin, Hua-Yi 3,4; Zhang, Xu-Yao 1,2; Zhu, Shunzhi 3,4
刊名	PATTERN RECOGNITION
	2023-08-01
卷号	140 页码:12
关键词	Chinese character recognition Zero-shot learning Self-information of radicals Character uncertainty elimination Radical information embedding
ISSN号	0031-3203
DOI	10.1016/j.patcog.2023.109598
通讯作者	Wang, Da-Han(wangdh@xmut.edu.cn)
英文摘要	Zero-shot Chinese character recognition (ZSCCR) is an important research topic in Chinese character recognition as it attempts to recognize unseen Chinese characters. As basic components and mid-level representations, radicals are significant for ZSCCR. However, previous methods treat the importance of radicals equally, ignoring the different contributions of radicals in distinguishing characters. In this pa-per, we propose the self-information of radicals (SIR) to measure the importance of radicals in recog-nizing Chinese characters. The proposed SIR can be easily adopted by two commonly used radical-based ZSCCR frameworks, i.e., sequence matching based and attribute embedding based. For sequence matching based ZSCCR, we propose a novel Chinese character uncertainty elimination (CUE) framework to allevi-ate the radical sequence mismatch problem. For attribute embedding based ZSCCR, we propose a novel radical information embedding (RIE) method that can highlight the importance of indispensable radi-cals and weaken the influence of some unnecessary radicals. We conducted comprehensive experiments on the CASIA-HWDB, ICDAR2013, CTW datasets, and AHCDB datasets to evaluate the proposed method. Experiments show that our proposed methods can achieve superior performance to the state-of-the-art methods, which demonstrate the effectiveness and the high extensibility of the proposed SIR.(c) 2023 Elsevier Ltd. All rights reserved.
资助项目	National Natural Science Foundation of China[61773325] ; National Natural Science Foundation of China[62222609] ; National Natural Science Foundation of China[62076236] ; Industry-University Cooperation Project of Fujian Science and Technology Department[2021H6035] ; Science and Technology Planning Project of Fujian Province[2021J011182] ; Science and Technology Planning Project of Fujian Province[2020H0023] ; Science and Technology Planning Project of Fujian Province[2020Y9064]
WOS关键词	STROKE EXTRACTION ; NETWORK
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	ELSEVIER SCI LTD
WOS记录号	WOS:000987191000001
资助机构	National Natural Science Foundation of China ; Industry-University Cooperation Project of Fujian Science and Technology Department ; Science and Technology Planning Project of Fujian Province
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/53290]
专题	多模态人工智能系统全国重点实验室
通讯作者	Wang, Da-Han
作者单位	1.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing, Peoples R China 2.Chinese Acad Sci, Natl Lab Pattern Recognit NLPR, Inst Automat, Beijing, Peoples R China 3.Fujian Key Lab Pattern Recognit & Image Understand, Xiamen, Peoples R China 4.Xiamen Univ Technol, Sch Comp & Informat Engineenng, Xiamen, Peoples R China
推荐引用方式 GB/T 7714	Luo, Guo-Feng,Wang, Da-Han,Du, Xia,et al. Self-information of radicals: A new clue for zero-shot Chinese character recognition[J]. PATTERN RECOGNITION,2023,140:12.
APA	Luo, Guo-Feng,Wang, Da-Han,Du, Xia,Yin, Hua-Yi,Zhang, Xu-Yao,&Zhu, Shunzhi.(2023).Self-information of radicals: A new clue for zero-shot Chinese character recognition.PATTERN RECOGNITION,140,12.
MLA	Luo, Guo-Feng,et al."Self-information of radicals: A new clue for zero-shot Chinese character recognition".PATTERN RECOGNITION 140(2023):12.