Neighbourhood preserving discriminant embedding in face recognition
Neighbourhood preserving discriminant embedding in face recognition
复制标题
DOI：
10.1016/j.jvcir.2009.08.003
复制
发表时间：
2009-11-01
影响因子：
2.6
通讯作者：
Abas, Fazly Salleh
中科院分区：
文献类型：
--
作者：
Han, Pang Ying;Jin, Andrew Teoh Beng;Abas, Fazly Salleh
本文提出了一种有效的用于人脸识别的鉴别特征提取方法。该技术结合了图嵌入和Fisher准则，我们称之为邻域保持判别嵌入（NPDE）。利用图嵌入准则，底层的非线性人脸数据结构被揭示为用于分析的代表性和区分性特征。我们使用邻域保持嵌入（NPE）来实现这一目的。NPE考虑了高维空间中的相邻点必须保持在低维空间中的相同邻域内并且位于类似的相对空间位置（而不改变每个数据点的最近邻居的局部结构）的限制。此外，利用Fisher准则的优点，NPDE的鉴别能力进一步提高。基于这种直觉，NPIDE获得了更好的区分能力，并在ORL，PIE和FRGC中得到了实验验证。(C)2009 Elsevier Inc. All rights reserved.
In this paper, we present an effective technique on discriminative feature extraction for face recognition. The proposed technique incorporates Graph Embedding and the Fisher's criterion where we call it as Neighbourhood Preserving Discriminant Embedding (NPDE). Utilizing the Graph Embedding criterion, the underlying nonlinear face data structure is revealed as representative and discriminative features for analysis. We employ Neighbourhood Preserving Embedding (NPE) for the purpose. NPE takes into account the restriction that neighbouring points in the high-dimensional space must remain within the same neighbourhood in the low dimension space and be located in a similar relative spatial situation (without changing the local structure of the nearest neighbours of each data point). Furthermore, by taking the advantage of the Fisher's criterion, the discriminating power of NPDE is further boosted. Based on this intuition, NPIDE obtains better discriminative capability and experimentally verified in ORL, PIE and FRGC. (C) 2009 Elsevier Inc. All rights reserved.