Abstract:Effective feature dimension reduction (DR) from high-dimensional remote sensing images has been a significant challenge for remote sensing object recognition. Directly adopting vector-based DR method ignores remote sensing data's inherent tensor structure information, leading to the undersample problem (USP). In addition, the existing tensor-based DR methods either require an exponential storage space increasing with the orders of the input tensor (i.e., Tucker-form methods) or are dependent on the permutation of tensor modes limiting the discriminant capability of the DR results (i.e., tensor train (TT) form methods). To conquer these problems, unlike the existing Tucker or TT form feature representation, the novel tensor ring (TR) subspace learning theory is proposed systematically and rigorously to extend the traditional vector and tensor subspace learning to the TR subspace. Then, by embedding the Fisher criterion into TR subspace, the TR discriminant analysis (TRDA) is proposed to achieve DR for remote sensing tensors with flexible tensor rank and lower storage cost. To train TRDA under different computing resources, nonrecursive and exact TRDA training methods are presented to obtain the global suboptimal and local optimal solutions, respectively. Furthermore, to adapt to the case of multisource data and unlabeled data, the multiple TRDA (MTRDA) and semi-supervised TRDA (S-TRDA) are further proposed to refine multisource features in multiple TR subspaces and absorb useful information using adaptive scatter tensor, respectively. Using optical, hyperspectral, and SAR datasets, experimental results demonstrate that the proposed TRDA can obtain better recognition accuracy and smaller storage cost than the typical vector and tensor-based DR methods.

Discriminant Neighborhood Structure Embedding Using Trace Ratio Criterion for Image Recognition

Two Dimension Linear Discriminant Analysis Based on Image Sampling and Regroupment

Combination of Local Multiple Patterns and Exponential Discriminant Analysis for facial recognition

Discriminant-enhanced neighborhood preserving embedding for dimensionality reduction

Two Dimensional Discriminant Neighborhood Preserving Embedding in Face Recognition

Semi-Supervised Dimension Reduction Using Trace Ratio Criterion.

Discriminant Neighborhood Embedding for Classification.

Direct Neighborhood Discriminant Analysis for Face Recognition

Tensor Rank Preserving Discriminant Analysis for Facial Recognition

An Effective Two-Dimensional Linear Discriminant Analysis with Locality Preserving Approach for Image Recognition

Neighborhood discriminant embedding in face recognition

Schatten-p Norm Based Linear Regression Discriminant Analysis for Face Recognition.

Tensor Ring Discriminant Analysis Used for Dimension Reduction of Remote Sensing Feature Tensor

Optimal Dimensionality Discriminant Analysis and Its Application to Image Recognition

Constrained discriminant neighborhood embedding for high dimensional data feature extraction.

Bilinear Neighborhood Discriminant Embedding And Its Non-Iterative Solution Algorithm

Beyond Trace Ratio: Weighted Harmonic Mean of Trace Ratios for Multiclass Discriminant Analysis

Discriminative Locally Linear Embedding in Image Space

Feature extraction using two-dimensional neighborhood margin and variation embedding

Trace Ratio Problem Revisited

Regularized Coplanar Discriminant Analysis for Dimensionality Reduction