ICPGAN: Intensity Component guided Pansharpening using Generative Adversarial Network with Dual Discriminators

CNN-based pansharpening methods use reduced resolution panchromatic and multispectral images due to the lack of a reference image, resulting in a mismatch problem when mapping to the reduced resolution images. We propose a pansharpening model which utilizes a reduced resolution multispectral image and the intensity component of a high resolution multispectral image instead of a reduced resolution panchromatic image, in the training process. The model comprises of two separate discriminators, each of which focuses on the spatial or spectral details of the given input. Additionally, the generator takes multispectral and panchromatic images, concatenates them and produces a synthetic image that closely resembles the original multispectral image. The results were compared to previous CNN-based methods and traditional methods both visually and in terms of evaluation metricd such as as ERGAS, SAM, QNR and Q.