GAF-NAU: Gramian Angular Field encoded Neighborhood Attention U-Net for Pixel-Wise Hyperspectral Image Classification

TL;DR

This paper introduces GAF-NAU, a novel pixel-based hyperspectral image classification method that transforms spectral data into 2D angular features and employs neighborhood attention within a U-Net architecture, outperforming patch-based approaches.

Contribution

The paper presents a new deep learning model that encodes spectral data with Gramian Angular Fields and uses neighborhood attention, avoiding patch assumptions in hyperspectral classification.

Findings

Superior accuracy on three public datasets

Effective suppression of irrelevant features

Outperforms existing patch-based methods

Abstract

Hyperspectral image (HSI) classification is the most vibrant area of research in the hyperspectral community due to the rich spectral information contained in HSI can greatly aid in identifying objects of interest. However, inherent non-linearity between materials and the corresponding spectral profiles brings two major challenges in HSI classification: interclass similarity and intraclass variability. Many advanced deep learning methods have attempted to address these issues from the perspective of a region/patch-based approach, instead of a pixel-based alternate. However, the patch-based approaches hypothesize that neighborhood pixels of a target pixel in a fixed spatial window belong to the same class. And this assumption is not always true. To address this problem, we herein propose a new deep learning architecture, namely Gramian Angular Field encoded Neighborhood Attention U-Net (GAF-NAU), for pixel-based HSI classification. The proposed method does not require regions or patches centered around a raw target pixel to perform 2D-CNN based classification, instead, our approach transforms 1D pixel vector in HSI into 2D angular feature space using Gramian Angular Field (GAF) and then embed it to a new neighborhood attention network to suppress irrelevant angular feature while emphasizing on pertinent features useful for HSI classification task. Evaluation results on three publicly available HSI datasets demonstrate the superior performance of the proposed model.