A 3D 2D convolutional Neural Network Model for Hyperspectral Image Classification

TL;DR

The paper introduces SEHybridSN, a 3D-2D convolutional neural network that effectively classifies hyperspectral images using minimal labeled data by leveraging hierarchical features and attention mechanisms.

Contribution

It proposes a novel hybrid CNN model with dense blocks, depthwise separable convolutions, and channel attention for improved hyperspectral image classification with limited training data.

Findings

Achieves high accuracy with only 0.05 and 0.01 labeled data

Utilizes hierarchical spatial spectral feature extraction

Demonstrates superior performance over existing methods

Abstract

In the proposed SEHybridSN model, a dense block was used to reuse shallow feature and aimed at better exploiting hierarchical spatial spectral feature. Subsequent depth separable convolutional layers were used to discriminate the spatial information. Further refinement of spatial spectral features was realized by the channel attention method, which were performed behind every 3D convolutional layer and every 2D convolutional layer. Experiment results indicate that our proposed model learn more discriminative spatial spectral features using very few training data. SEHybridSN using only 0.05 and 0.01 labeled data for training, a very satisfactory performance is obtained.