Unsupervised Denoising for Satellite Imagery using Wavelet Subband CycleGAN
This addresses a domain-specific issue for geological survey applications by providing an unsupervised solution to improve satellite image quality, though it is incremental as it builds on existing CycleGAN and wavelet techniques.
The paper tackles the problem of removing sensor noises like vertical stripes and waves from multi-spectral satellite imagery, which is challenging due to the lack of paired noisy-clean data, by proposing an unsupervised method using wavelet subband CycleGAN that effectively denoises while preserving high-frequency features.
Multi-spectral satellite imaging sensors acquire various spectral band images such as red (R), green (G), blue (B), near-infrared (N), etc. Thanks to the unique spectroscopic property of each spectral band with respective to the objects on the ground, multi-spectral satellite imagery can be used for various geological survey applications. Unfortunately, image artifacts from imaging sensor noises often affect the quality of scenes and have negative impacts on the applications of satellite imagery. Recently, deep learning approaches have been extensively explored for the removal of noises in satellite imagery. Most deep learning denoising methods, however, follow a supervised learning scheme, which requires matched noisy image and clean image pairs that are difficult to collect in real situations. In this paper, we propose a novel unsupervised multispectral denoising method for satellite imagery using wavelet subband cycle-consistent adversarial network (WavCycleGAN). The proposed method is based on unsupervised learning scheme using adversarial loss and cycle-consistency loss to overcome the lack of paired data. Moreover, in contrast to the standard image domain cycleGAN, we introduce a wavelet subband domain learning scheme for effective denoising without sacrificing high frequency components such as edges and detail information. Experimental results for the removal of vertical stripe and wave noises in satellite imaging sensors demonstrate that the proposed method effectively removes noises and preserves important high frequency features of satellite images.