Evaluation of weighted Fisher criteria for large category dimensionality reduction in application to Chinese handwriting recognition

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

	Evaluation of weighted Fisher criteria for large category dimensionality reduction in application to Chinese handwriting recognition
	Zhang, Xu-Yao; Liu, Cheng-Lin
发表期刊	PATTERN RECOGNITION
	2013-09-01
卷号	46 期号:9 页码:2599-2611
文章类型	Article
摘要	To improve the class separability of Fisher linear discriminant analysis (FDA) for large category problems, we investigate the weighted Fisher criterion (WFC) by integrating weighting functions for dimensionality reduction. The objective of WFC is to maximize the sum of weighted distances of all class pairs. By setting larger weights for the most confusable classes, WFC can improve the class separation while the solution remains an eigen-decomposition problem. We evaluate five weighting functions in three different weighting spaces in a typical large category problem of handwritten Chinese character recognition. The weighting functions include four based on existing methods, namely, FDA, approximate pairwise accuracy criterion (aPAC), power function (POW), confused distance maximization (CDM), and a new one based on K-nearest neighbors (KNN). All the weighting functions can be calculated in the original feature space, low-dimensional space, or fractional space. Our experiments on a 3,755-class Chinese handwriting database demonstrate that WFC can improve the classification accuracy significantly compared to FDA. Among the weighting functions, the KNN method in the original space is the most competitive model which achieves significantly higher classification accuracy and has a low computational complexity. To further improve the performance, we propose a nonparametric extension of the KNN method from the class level to the sample level. The sample level KNN (SKNN) method is shown to outperform significantly other methods in Chinese handwriting recognition such as the locally linear discriminant analysis (LLDA), neighbor class linear discriminant analysis (NCLDA), and heteroscedastic linear discriminant analysis (HLDA). (C) 2013 Elsevier Ltd. All rights reserved.
关键词	Dimensionality Reduction Large Category Class Separation Problem Weighted Fisher Criteria Class Level Sample Level Chinese Handwriting Recognition
WOS标题词	Science & Technology ; Technology
关键词[WOS]	CHARACTER-RECOGNITION ; FEATURE-EXTRACTION ; DISTANCE MAXIMIZATION ; DISCRIMINANT-ANALYSIS ; COMPONENT ANALYSIS ; FACE RECOGNITION ; LDA ; FRAMEWORK
收录类别	SCI
语种	英语
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号	WOS:000318837400017
引用统计	被引频次：17[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/173211/3090
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	Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, 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	Zhang, Xu-Yao,Liu, Cheng-Lin. Evaluation of weighted Fisher criteria for large category dimensionality reduction in application to Chinese handwriting recognition[J]. PATTERN RECOGNITION,2013,46(9):2599-2611.
APA	Zhang, Xu-Yao,&Liu, Cheng-Lin.(2013).Evaluation of weighted Fisher criteria for large category dimensionality reduction in application to Chinese handwriting recognition.PATTERN RECOGNITION,46(9),2599-2611.
MLA	Zhang, Xu-Yao,et al."Evaluation of weighted Fisher criteria for large category dimensionality reduction in application to Chinese handwriting recognition".PATTERN RECOGNITION 46.9(2013):2599-2611.

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
XYZ2013-weighted_Fis（1095KB）	期刊论文	作者接受稿	开放获取	CC BY-NC-SA	浏览下载