Spatial Context based Angular Information Preserving Projection for Hyperspectral Image Classification

TL;DR

This paper introduces an unsupervised spatially-aware angular information preserving projection method for hyperspectral image classification, enhancing local spatial and angular features to improve classification accuracy.

Contribution

It proposes a novel unsupervised dimensionality reduction technique that incorporates spatial context, specifically designed for hyperspectral data, and a sparse classifier optimized for spatial information.

Findings

Improved classification accuracy on real-world datasets.

Effective preservation of local angular and spatial information.

Robustness against variations in hyperspectral data.

Abstract

Dimensionality reduction is a crucial preprocessing for hyperspectral data analysis - finding an appropriate subspace is often required for subsequent image classification. In recent work, we proposed supervised angular information based dimensionality reduction methods to find effective subspaces. Since unlabeled data are often more readily available compared to labeled data, we propose an unsupervised projection that finds a lower dimensional subspace where local angular information is preserved. To exploit spatial information from the hyperspectral images, we further extend our unsupervised projection to incorporate spatial contextual information around each pixel in the image. Additionally, we also propose a sparse representation based classifier which is optimized to exploit spatial information during classification - we hence assert that our proposed projection is particularly suitable for classifiers where local similarity and spatial context are both important. Experimental results with two real-world hyperspectral datasets demonstrate that our proposed methods provide a robust classification performance.