您的位置:首页  > 论文页面

规则化的线性鉴别分析算法

发表时间:2011-04-30  浏览量:1598  下载量:635
全部作者: 郭礼华,金连文
作者单位: 华南理工大学电子与信息学院
摘 要: 针对线性鉴别分析(linear discriminant analysis,LDA)的多模态问题,提出一种改进型LDA(improved linear discriminant analysis, ILDA)算法。此算法借鉴规则化的思想,重新定义类内散度矩阵和类间散度矩阵,同时针对实际问题,通过判断矩阵迹大小优化求解LDA投影矩阵。实验结果表明:ILDA算法不仅保留了LDA选择鉴别子空间的性能,还可以很好地解决某些多模态问题。
关 键 词: 图像处理;线性鉴别分析;规则化;多模态
Title: An improve linear discriminant analysis method based on regularization
Author: GUO Lihua, JIN Lianwen
Organization: School of Electronic and Information Engineering, South China University of Technology
Abstract: Based on linear discriminant analysis (LDA) method, this paper proposes an improved LDA (ILDA) method to overcome the multi-model problem of LDA. In the ILDA method, the between-class scatter matrix and within-class scatter matrix are regularized, and some rules are introduced to optimize the Eigen analysis of LDA using matrix trace judgment. Finally, some experimental results show that ILDA method can preserve the ability to choose the discriminant low-dimension subspace, and overcome some multi-model problems.
Key words: image processing; linear discriminant analysis; regularization; multi-model problem
发表期数: 2011年4月第8期
引用格式: 郭礼华,金连文. 规则化的线性鉴别分析算法[J]. 中国科技论文在线精品论文,2011,4(8):755-759.
 
0 评论数 0
暂无评论
友情链接