HyperDID: Hyperspectral Intrinsic Image Decomposition with Deep Feature Embedding

TL;DR

HyperDID introduces a deep feature embedding framework with specialized modules to enhance hyperspectral intrinsic image decomposition, significantly improving classification accuracy across multiple datasets.

Contribution

It proposes a novel deep feature embedding approach with dedicated modules for intrinsic feature extraction and separation, advancing hyperspectral image classification.

Findings

Improved classification accuracy on three datasets

Effective separation of environment and category features

Validated robustness of the proposed method

Abstract

The dissection of hyperspectral images into intrinsic components through hyperspectral intrinsic image decomposition (HIID) enhances the interpretability of hyperspectral data, providing a foundation for more accurate classification outcomes. However, the classification performance of HIID is constrained by the model's representational ability. To address this limitation, this study rethinks hyperspectral intrinsic image decomposition for classification tasks by introducing deep feature embedding. The proposed framework, HyperDID, incorporates the Environmental Feature Module (EFM) and Categorical Feature Module (CFM) to extract intrinsic features. Additionally, a Feature Discrimination Module (FDM) is introduced to separate environment-related and category-related features. Experimental results across three commonly used datasets validate the effectiveness of HyperDID in improving hyperspectral image classification performance. This novel approach holds promise for advancing the capabilities of hyperspectral image analysis by leveraging deep feature embedding principles. The implementation of the proposed method could be accessed soon at https://github.com/shendu-sw/HyperDID for the sake of reproducibility.