Improving handwritten Chinese text recognition using neural network language models and convolutional neural network shape models

doi:10.1016/j.patcog.2016.12.026

CASIA OpenIR > 多模态人工智能系统全国重点实验室 > 模式分析与学习

	Improving handwritten Chinese text recognition using neural network language models and convolutional neural network shape models
	Wu, Yi-Chao1 ; Yin, Fei1 ; Liu, Cheng-Lin1,2,3
发表期刊	PATTERN RECOGNITION
	2017-05-01
卷号	2017 期号:65 页码:251-264
文章类型	Article
摘要	Handwritten Chinese text recognition based on over-segmentation and path search integrating multiple contexts has been demonstrated successful, wherein the language model (LM) and character shape models play important roles. Although back-off N-gram LMs (BLMs) have been used dominantly for decades, they suffer from the data sparseness problem, especially for high-order LMs. Recently, neural network LMs (NNLMs) have been applied to handwriting recognition with superiority to BLMs. With the aim of improving Chinese handwriting recognition, this paper evaluates the effects of two types of character-level NNLMs, namely, feedforward neural network LMs (FNNLMs) and recurrent neural network LMs (RNNLMs). Both FNNLMs and RNNLMs are also combined with BLMs to construct hybrid LMs. For fair comparison with BLMs and a state-of-the-art system, we evaluate in a system with the same character over-segmentation and classification techniques as before, and compare various LMs using a small text corpus used before. Experimental results on the Chinese handwriting database CASIA-HWDB validate that NNLMs improve the recognition performance, and hybrid RNNLMs outperform the other LMs. To report a new benchmark, we also evaluate selected LMs on a large corpus, and replace the baseline character classifier, over-segmentation, and geometric context models with convolutional neural network (CNN) based models. The performance on both the CASIA-HWDB and the ICDAR-2013 competition dataset are improved significantly. On the CASIA-HWDB test set, the character-level accurate rate (AR) and correct rate (CR) achieve 95.88% and 95.95%, respectively.
关键词	Handwritten Chinese Text Recognition Feedforward Neural Network Language Model Recurrent Neural Network Language Model Hybrid Language Model Convolutional Neural Network Shape Models
WOS标题词	Science & Technology ; Technology
DOI	10.1016/j.patcog.2016.12.026
关键词[WOS]	CHARACTER-RECOGNITION ; DOCUMENT RECOGNITION ; SEGMENTATION ; STRINGS ; ONLINE
收录类别	SCI
语种	英语
项目资助者	National Natural Science Foundation of China (NSFC)(61305005 ; 61273269 ; 61573355 ; 61411136002)
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号	WOS:000394197700021
引用统计	被引频次：71[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/13428
专题	多模态人工智能系统全国重点实验室_模式分析与学习
作者单位	1.Chinese Acad Sci, Inst Inst Automat, NLPR, 95 Zhongguancun East Rd, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing, Peoples R China 3.Chinese Acad Sci, Ctr Excellence Brain Sci & Intelligence Technol, Beijing, Peoples R China
第一作者单位	模式识别国家重点实验室
推荐引用方式 GB/T 7714	Wu, Yi-Chao,Yin, Fei,Liu, Cheng-Lin. Improving handwritten Chinese text recognition using neural network language models and convolutional neural network shape models[J]. PATTERN RECOGNITION,2017,2017(65):251-264.
APA	Wu, Yi-Chao,Yin, Fei,&Liu, Cheng-Lin.(2017).Improving handwritten Chinese text recognition using neural network language models and convolutional neural network shape models.PATTERN RECOGNITION,2017(65),251-264.
MLA	Wu, Yi-Chao,et al."Improving handwritten Chinese text recognition using neural network language models and convolutional neural network shape models".PATTERN RECOGNITION 2017.65(2017):251-264.

条目包含的文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
1-s2.0-S003132031630（1290KB）	期刊论文	作者接受稿	开放获取	CC BY-NC-SA	浏览下载