Learning a Representation with the Block-Diagonal Structure for Pattern Classification

TL;DR

This paper introduces RBDS, a novel method that learns a block-diagonal structured representation to improve robustness in image classification, especially under corrupted data conditions.

Contribution

The paper proposes a new regularization approach that enforces block-diagonal structure in representations, enhancing robustness of SRC-based classification.

Findings

RBDS improves classification accuracy on benchmark datasets.

The method effectively handles corrupted training and test data.

Experimental results validate the robustness of the proposed approach.

Abstract

Sparse-representation-based classification (SRC) has been widely studied and developed for various practical signal classification applications. However, the performance of a SRC-based method is degraded when both the training and test data are corrupted. To counteract this problem, we propose an approach that learns Representation with Block-Diagonal Structure (RBDS) for robust image recognition. To be more specific, we first introduce a regularization term that captures the block-diagonal structure of the target representation matrix of the training data. The resulting problem is then solved by an optimizer. Last, based on the learned representation, a simple yet effective linear classifier is used for the classification task. The experimental results obtained on several benchmarking datasets demonstrate the efficacy of the proposed RBDS method.