Exploiting Clustering Manifold Structure for Hyperspectral Imagery Super-Resolution.

Area of Science:

• Remote Sensing
• Computer Vision
• Image Processing

Background:

• Hyperspectral image (HSI) super-resolution aims to fuse low-resolution HSI with high-resolution (HR) conventional images.
• Previous methods often overlook the spatial manifold structure of the latent HR HSI.

Purpose of the Study:

• To develop a novel HSI super-resolution method by exploiting the spatial manifold structure of the latent HR HSI.
• To improve the accuracy and performance of HSI super-resolution.

Main Methods:

• A clustering approach is applied to the spatial domain of the input conventional image.
• An intra-cluster self-expressiveness model is used to depict the clustering manifold structure.
• The learned structure is integrated into a variational super-resolution framework, optimized via an alternating direction method of multipliers.

Main Results:

• The proposed method effectively preserves the underlying spatial manifold structure of the latent HSI.
• Demonstrated state-of-the-art super-resolution performance on two benchmark datasets.
• The method successfully enhances the spatial resolution of hyperspectral images.

Conclusions:

• Exploiting the spatial manifold structure is crucial for advanced HSI super-resolution.
• The proposed method offers a robust and effective solution for generating high-resolution hyperspectral images.
• This work advances the field of hyperspectral imaging and its applications.