A Self-Supervised Denoising Network for Satellite-Airborne-Ground Hyperspectral Imagery
Xinyu Wang,Zhaozhi Luo,Wenqing Li,Xin Hu,Liangpei Zhang,Yanfei Zhong
DOI: https://doi.org/10.1109/tgrs.2021.3064429
IF: 8.2
2021-01-01
IEEE Transactions on Geoscience and Remote Sensing
Abstract:Hyperspectral images (HSIs) are inevitably corrupted with various types of noise, which seriously degrades the data quality and usability. Denoising is an essential preprocessing task of HSI processing. Recently, benefiting from the great learning ability of deep learning, convolutional neural network (CNN) denoisers have obtained state-of-the-art performances for Gaussian noise removal. However, one central problem remains largely unsolved: how to deal with the complicated noise in the real-world HSIs, especially when a paired training data set is unavailable. In this article, a self-supervised hyperspectral image denoising network (SHDN) is proposed, which consists of a noise estimator and a CNN denoiser. Rather than defining a complex noise model to generate training pairs on the clean HSIs, a self-supervised training scheme is first proposed by considering the noisy HSI itself as the training data. Through the noise estimator, the realistic noise samples can be extracted and combined with the clean bands to make up the training pairs. In addition, to jointly restore the target noisy band and to maintain the spectral consistency, a flexible multi-to-single band convolutional network is designed, where the noisy band and the neighboring bands are jointly aggregated via multiscale contextualized dilated blocks and the spectral–spatial convolutional unit. Experiments on HSIs from spaceborne, airborne, unmanned aerial vehicle (UAV)-borne, and ground-based data sets demonstrate the applicability and the generalization of SHDN in the real scenarios. Additionally, the usability of the noisy bands and the suitability of the SHDN framework in the subsequent applications are verified in the land-cover mapping experiments.