Bilateral Two-Dimensional Locality Preserving Projections

Chen, Sibao; Luo, Bin; Hu, Guoping; Wang, Ren-Hua

doi:10.1109/icassp.2007.366307

Cited by 8 publications

(3 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The objective functions in (21) and (22) indicate that the proposed B2DMRPDE is a supervised learning method because it takes the class label information into consideration and constructs the within-reconstruction and the between-reconstruction graphs.…”

Section: B the Objective Functionmentioning

confidence: 99%

“…Song et al [19] proposed a bilateral two-dimensional locality preserving projections (B2DLPP) algorithm, which effectively reduces the dimensionality from the horizontal and vertical directions of the data matrix. Other work related to B2DLPP can be seen in [20] and [21]. Similar to B2DLPP, Zhang et al [22] proposed a bilateral two-dimensional neighborhood preserving embedding (B2DNPE) algorithm.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Bilateral Two-Dimensional Matrix Regression Preserving Discriminant Embedding for Corrupted Image Recognition

Zhang

Wang

2019

IEEE Access

View full text Add to dashboard Cite

Nuclear-norm-based matrix regression (NMR) methods have been successfully applied for the recognition of corrupted images. However, most of these methods do not consider the label information and are classified as unsupervised learning methods. In this paper, we propose a new regression-based algorithm, named bilateral two-dimensional matrix regression preserving discriminant embedding (B2DMRPDE). The proposed algorithm constructs the within-reconstruction graph and between-reconstruction graph using NMR. Then, B2DMRPDE aims to seek a subspace in which the within-class reconstructive residual is minimized and the between-class reconstructive residual is maximized based on Fisher's criterion. Hence, B2DMRPDE can capture the potential discriminative information for classification. To enhance the classification effectiveness, we present a new NMR-based classifier to determine the class label of the testing sample. Extensive experiments on face image databases were performed, and the results validate the effectiveness of the proposed method. INDEX TERMS Corrupted image, face recognition, low-rank, matrix regression, nuclear-norm.

show abstract

Section: B the Objective Functionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Bilateral Two-Dimensional Matrix Regression Preserving Discriminant Embedding for Corrupted Image Recognition

Zhang

Wang

2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…In fact, two dimensional methods perform compression only in row direction. Bilateral compression methods, such as ( ) 2 2D PCA [12], ( ) 2 2D FLD [13] and B2DLPP [14], perform compression not only in row direction, but also in column direction. So, image matrices can be represented by fewer coefficients.…”

Section: Introductionmentioning

confidence: 99%

Two-Dimensional Extensions of Neighborhood Preserving Embedding

Zhang

Liu

Chen³

et al. 2012

AMM

View full text Add to dashboard Cite

Neighbourhood Preserving Embedding (NPE) is a novel subspace learning algorithm, which aims at preserving the local neighbourhood structure on the data manifold and is a linear approximation to Locally Linear Embedding (LLE). However, in typical image recognition in 1D vectors space, where the number of data samples is smaller than the dimension of data space, suffering from the singularity problem of matrix, NPE algorithm cannot be implemented directly. In this paper, we investigate NPE directly on image matrix for image recognition. The proposed two-dimensional neighbourhood preserving embedding (2DNPE) and bilateral two-dimensional neighbourhood preserving embedding (B2DNPE) algorithms are all based directly on 2D image matrices rather than on 1D vectors as NPE does, thus the problem of singularity confronted in 1D case is overcome. 2DNPE performs compression only in row direction, while B2DNPE performs compression both in row and in column direction. The relation of them to 2DLPP (B2DLPP) are also presented. The proposed algorithms are evaluated on ORL face database and handwritten digits database.

show abstract