A Hybrid CNN Based on Global Reasoning for Hyperspectral Image Classification

In recent years, convolutional neural networks (CNNs) have been widely used in hyperspectral images (HSIs) classification. However, 2-D CNN, 3-D CNN, and even the newly emerged hybrid CNN (HCNN) all require multiple or deep CNN layers to obtain excellent classification performance, which inevitably results in the high complexity and the need for a large number of training samples. Moreover, as a local operator, convolution is challenging to fully use global information. To solve the above two issues, we design a HCNN based on global reasoning (GloRe-HCNN) for HSI classification. On the one hand, the GloRe-HCNN uses only one layer of 3-D CNN and one layer of 2-D CNN to jointly extract the spatial–spectral features of HSI. On the other hand, we contrive a spatial–spectral global reasoning unit (SS-GloRe-Unit) to take the place of stacked multilayer 3-D CNN for extracting global features fully. We select small training samples in three standard datasets and compare them with state-of-the-art CNN methods. Numerous experiments show that our GloRe-HCNN performs advanced performance.

I. Introduction

Hyperspectral image (HSI) offers excellent help to the extraction of ground object information because of many spectral bands. However, a large number of spectral bands also causes the redundancy of ground object information and the higher complexity of data processing, which brings challenges to HSI classification. Especially when the training sample is small, it will produce more difficulties in HSI classification. Currently, classification methods of HSI mainly fall into two categories: traditional methods and deep learning methods [1].