Incremental threshold learning for classifier selection

CORC > 西安光学精密机械研究所 > 中国科学院西安光学精密机械研究所 > 光学影像学习与分析中心

	Incremental threshold learning for classifier selection
	Pang, Yanwei 2; Deng, Junping 2; Yuan, Yuan1
刊名	neurocomputing
	2012-07-15
卷号	89 页码:89-95
关键词	Incremental learning Threshold-based classifier Classifier fusion Object detection Pattern recognition
ISSN号	0925-2312
产权排序	2
合作状况	国内
中文摘要	threshold-based classifier is a simple yet powerful pattern classification tool, which has been frequently used in applications of object detection and recognition. a threshold-based classifier is usually associated with a unique one-dimensional feature. a properly selected threshold and a binary sign corresponding to the feature govern the classifier. however, the learning process is usually done in a batch manner. the batch algorithms are not suitable for sequentially incoming data because of the limitation of storage and prohibitive computation cost. to deal with sequentially incoming data, this paper proposes an incremental algorithm for incrementally learning the threshold-based classifiers. the proposed method can not only incrementally model the features but also estimate the threshold and training error in a close form. the effectiveness of the proposed algorithm is evaluated in the applications of gender recognition, face detection, and human detection. (c) 2012 elsevier b.v. all rights reserved.
英文摘要	threshold-based classifier is a simple yet powerful pattern classification tool, which has been frequently used in applications of object detection and recognition. a threshold-based classifier is usually associated with a unique one-dimensional feature. a properly selected threshold and a binary sign corresponding to the feature govern the classifier. however, the learning process is usually done in a batch manner. the batch algorithms are not suitable for sequentially incoming data because of the limitation of storage and prohibitive computation cost. to deal with sequentially incoming data, this paper proposes an incremental algorithm for incrementally learning the threshold-based classifiers. the proposed method can not only incrementally model the features but also estimate the threshold and training error in a close form. the effectiveness of the proposed algorithm is evaluated in the applications of gender recognition, face detection, and human detection. (c) 2012 elsevier b.v. all rights reserved.
学科主题	computer science ; artificial intelligence
WOS标题词	science & technology ; technology
类目[WOS]	computer science, artificial intelligence
研究领域[WOS]	computer science
关键词[WOS]	local binary patterns ; face-recognition ; neural-network ; propagation ; scale
收录类别	SCI ; EI
语种	英语
WOS记录号	WOS:000304638500009
公开日期	2012-09-03
内容类型	期刊论文
源URL	[http://ir.opt.ac.cn/handle/181661/20259]
专题	西安光学精密机械研究所_光学影像学习与分析中心
作者单位	1.Chinese Acad Sci, Xian Inst Opt & Precis Mech, State Key Lab Transient Opt & Photon, Xian 710119, Peoples R China 2.Tianjin Univ, Sch Elect Informat Engn, Tianjin 300072, Peoples R China
推荐引用方式 GB/T 7714	Pang, Yanwei,Deng, Junping,Yuan, Yuan. Incremental threshold learning for classifier selection[J]. neurocomputing,2012,89:89-95.
APA	Pang, Yanwei,Deng, Junping,&Yuan, Yuan.(2012).Incremental threshold learning for classifier selection.neurocomputing,89,89-95.
MLA	Pang, Yanwei,et al."Incremental threshold learning for classifier selection".neurocomputing 89(2012):89-95.