An Improved Method for Pan-Sharpening Based on Pan-GAN

Yan Li,Jianmin Li,Xiaofeng Du,Yibo Huang,Jian Lei
DOI: https://doi.org/10.1109/ICIVC55077.2022.9887169
2022-01-01
Abstract:Pan-sharpening refers to the fusion of pan images and low-resolution multispectral remote sensing images to obtain high-resolution multispectral images. Generative adversarial network (GAN)-based pan-sharpening methods have recently became popular due to the lack of ground-truth data during training. However, GAN-based methods suffer from training instability and convergence difficulties. To deal the issues, we propose a novel GAN-based pan-sharpening method using additional constraint. First, we adopt the geometric consistency constraint to enforce the network to preserve the spatial structure of image. Second, we introduce an attention mechanism in the generator to extract useful information through the features of pan image and multispectral images and pay more attention to meaningful regions. Experimental results show the effectiveness of our method in terms of the quantitative and visual results.
What problem does this paper attempt to address?