SSCAN: A Spatial–Spectral Cross Attention Network for Hyperspectral Image Denoising

Hyperspectral images (HSIs) have been widely used in a variety of applications thanks to the rich spectral information they are able to provide. Among all HSI processing tasks, HSI denoising is a crucial step. Recent years have seen great progress in deep learning-based image denoising methods. However, existing efforts tend to ignore the correlations between adjacent spectral bands, leading to problems such as spectral distortion and blurred edges in denoised results. In this study, we propose a novel HSI denoising network, termed spectral–spatial cross attention network (SSCAN), that combines group convolutions and attention modules. Specifically, we use a group convolution with a spatial attention module to facilitate feature extraction by directing models’ attention to bandwise important features. We also propose a spectral–spatial attention block (SSAB) to effectively exploit the spatial and spectral information in HSIs. In addition, we adopt residual learning operations with skip connections to ensure training stability. The experimental results indicate that the proposed SSCAN outperforms several state-of-the-art HSI denoising algorithms.

I. Introduction

Different from traditional RGB images, hyperspectral images (HSIs) contain more spatial and spectral information that benefits a variety of remote sensing applications, such as target detection, classifications, and reconnaissance. The last decade has seen rapid development of the HSI acquisition approaches [1]. However, due to reasons that include atmospheric interference and sensor restrictions, HSIs can be corrupted by several types of noises, including Gaussian noises, stripe noises, and impulse noises, which largely degrade the image quality and affect subsequent applications. Therefore, denoising is considered an essential preprocessing step for HSI applications.