Hyperspectral Classification Based on 3D Asymmetric Inception Network with Data Fusion Transfer Learning

TL;DR

This paper introduces AINet, a 3D asymmetric inception network with a data fusion transfer learning strategy, significantly improving hyperspectral image classification accuracy on multiple benchmarks by addressing overfitting and spectral-spatial feature extraction.

Contribution

The paper proposes a novel 3D asymmetric inception network and a data fusion transfer learning approach to enhance hyperspectral image classification performance.

Findings

AINet outperforms state-of-the-art methods on multiple benchmarks

The transfer learning strategy boosts classification accuracy

AINet effectively captures spectral signatures over spatial context

Abstract

Hyperspectral image(HSI) classification has been improved with convolutional neural network(CNN) in very recent years. Being different from the RGB datasets, different HSI datasets are generally captured by various remote sensors and have different spectral configurations. Moreover, each HSI dataset only contains very limited training samples and thus it is prone to overfitting when using deep CNNs. In this paper, we first deliver a 3D asymmetric inception network, AINet, to overcome the overfitting problem. With the emphasis on spectral signatures over spatial contexts of HSI data, AINet can convey and classify the features effectively. In addition, the proposed data fusion transfer learning strategy is beneficial in boosting the classification performance. Extensive experiments show that the proposed approach beat all of the state-of-art methods on several HSI benchmarks, including Pavia University, Indian Pines and Kennedy Space Center(KSC). Code can be found at: https://github.com/UniLauX/AINet.