Multi-order co-occurrence activations encoded with Fisher Vector for scene character recognition

doi:10.1016/j.patrec.2017.07.011

CASIA OpenIR > 复杂系统管理与控制国家重点实验室 > 影像分析与机器视觉

	Multi-order co-occurrence activations encoded with Fisher Vector for scene character recognition
	Wang, Yanna1,2 ; Shi, Cunzhao1,2 ; Wang, Chunheng1,2 ; Xiao, Baihua1,2 ; Qi, Chengzuo1,2
发表期刊	PATTERN RECOGNITION LETTERS
	2017-10-01
卷号	97 期号:2017 页码:69-76
文章类型	Article
摘要	Scene character recognition remains a challenging task due to many interference factors. Considering that characters are composed of a series of parts arranged in certain structures, in this paper, we propose a novel representation termed multi-order co-occurrence activations (MCA) encoded with Fisher Vector (FV), namely MCA-FV. It implicitly models the co-occurrence information of discriminative character parts at different orders to boost the recognition performance. We first extract convolutional activations as local descriptors of character parts from convolutional neural networks (CNNs). Then, we introduce MCA features to capture the multi-order co-occurrence cues among different discriminative character parts. Finally, we apply FV to encode co-occurrence features of each order and obtain a global representation of MCA-FV. The proposed method is evaluated on four scene character datasets including English and Chinese datasets. Experiment results demonstrate the effectiveness of the proposed method for scene character recognition. (C) 2017 Elsevier B.V. All rights reserved.
关键词	Scene Character Recognition Multi-order Co-occurrence Activations Fisher Vector Cnn
WOS标题词	Science & Technology ; Technology
DOI	10.1016/j.patrec.2017.07.011
关键词[WOS]	TEXT RECOGNITION ; ORIENTED GRADIENTS ; REPRESENTATION ; STROKELETS ; HISTOGRAM
收录类别	SCI
语种	英语
项目资助者	National Natural Science Foundation of China (NSFC)(61601462 ; 61531019 ; 71621002)
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000411765800011
引用统计	被引频次：14[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/173211/19587
专题	复杂系统管理与控制国家重点实验室_影像分析与机器视觉
作者单位	1.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing, Peoples R China
第一作者单位	中国科学院自动化研究所
推荐引用方式 GB/T 7714	Wang, Yanna,Shi, Cunzhao,Wang, Chunheng,et al. Multi-order co-occurrence activations encoded with Fisher Vector for scene character recognition[J]. PATTERN RECOGNITION LETTERS,2017,97(2017):69-76.
APA	Wang, Yanna,Shi, Cunzhao,Wang, Chunheng,Xiao, Baihua,&Qi, Chengzuo.(2017).Multi-order co-occurrence activations encoded with Fisher Vector for scene character recognition.PATTERN RECOGNITION LETTERS,97(2017),69-76.
MLA	Wang, Yanna,et al."Multi-order co-occurrence activations encoded with Fisher Vector for scene character recognition".PATTERN RECOGNITION LETTERS 97.2017(2017):69-76.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
Multi-order co-occur（1105KB）	期刊论文	作者接受稿	开放获取	CC BY-NC-SA	浏览下载