3D-ANAS: 3D Asymmetric Neural Architecture Search for Fast Hyperspectral Image Classification

TL;DR

This paper introduces 3D-ANAS, an automated neural architecture search method for hyperspectral image classification, which designs efficient 3D asymmetric networks and a fast pixel-to-pixel framework, achieving high accuracy and speed.

Contribution

The paper proposes a novel 3D asymmetric neural architecture search algorithm and a fast pixel-to-pixel classification framework for hyperspectral images, reducing manual design effort and computational costs.

Findings

Achieves competitive accuracy on multiple hyperspectral datasets.

Significantly faster inference speed compared to state-of-the-art methods.

Reduces computational and design complexity in hyperspectral image classification.

Abstract

Hyperspectral images involve abundant spectral and spatial information, playing an irreplaceable role in land-cover classification. Recently, based on deep learning technologies, an increasing number of HSI classification approaches have been proposed, which demonstrate promising performance. However, previous studies suffer from two major drawbacks: 1) the architecture of most deep learning models is manually designed, relies on specialized knowledge, and is relatively tedious. Moreover, in HSI classifications, datasets captured by different sensors have different physical properties. Correspondingly, different models need to be designed for different datasets, which further increases the workload of designing architectures; 2) the mainstream framework is a patch-to-pixel framework. The overlap regions of patches of adjacent pixels are calculated repeatedly, which increases computational cost and time cost. Besides, the classification accuracy is sensitive to the patch size, which is artificially set based on extensive investigation experiments. To overcome the issues mentioned above, we firstly propose a 3D asymmetric neural network search algorithm and leverage it to automatically search for efficient architectures for HSI classifications. By analysing the characteristics of HSIs, we specifically build a 3D asymmetric decomposition search space, where spectral and spatial information are processed with different decomposition convolutions. Furthermore, we propose a new fast classification framework, i,e., pixel-to-pixel classification framework, which has no repetitive operations and reduces the overall cost. Experiments on three public HSI datasets captured by different sensors demonstrate the networks designed by our 3D-ANAS achieve competitive performance compared to several state-of-the-art methods, while having a much faster inference speed.