Fuzzy local maximal marginal embedding for feature extraction

Authors:
Cairong Zhao;Zhihui Lai;Chuancai Liu;Xingjian Gu;Jianjun Qian
Affiliations:
Nanjing University of Science and Technology, School of Computer Science, 210094, Nanjing, Jiansu, China and Minjiang College, Department of Physics and Electronics, 350108, Fuzhou, Fujian, China;Nanjing University of Science and Technology, School of Computer Science, 210094, Nanjing, Jiansu, China;Nanjing University of Science and Technology, School of Computer Science, 210094, Nanjing, Jiansu, China;Nanjing University of Science and Technology, School of Computer Science, 210094, Nanjing, Jiansu, China;Nanjing University of Science and Technology, School of Computer Science, 210094, Nanjing, Jiansu, China
Venue:
Soft Computing - A Fusion of Foundations, Methodologies and Applications
Year:
2012

Citing 0
Cited 2

Two-dimensional color uncorrelated discriminant analysis for face recognition

Neurocomputing
Scene classification using multi-resolution low-level feature combination

Neurocomputing

Quantified Score

Hi-index	0.01

Visualization

Abstract

In graph-based linear dimensionality reduction algorithms, it is crucial to construct a neighbor graph that can correctly reflect the relationship between samples. This paper presents an improved algorithm called fuzzy local maximal marginal embedding (FLMME) for linear dimensionality reduction. Significantly differing from the existing graph-based algorithms is that two novel fuzzy gradual graphs are constructed in FLMME, which help to pull the near neighbor samples in same class nearer and nearer and repel the far neighbor samples of margin between different classes farther and farther when they are projected to feature subspace. Through the fuzzy gradual graphs, FLMME algorithm has lower sensitivities to the sample variations caused by varying illumination, expression, viewing conditions and shapes. The proposed FLMME algorithm is evaluated through experiments by using the WINE database, the Yale and ORL face image databases and the USPS handwriting digital databases. The results show that the FLMME outperforms PCA, LDA, LPP and local maximal marginal embedding.