Target-adaptive CNN-based pansharpening
This is an incremental improvement for remote sensing users, enabling fast and high-quality pansharpening on general-purpose hardware.
The paper tackles remote sensing image pansharpening by proposing a target-adaptive CNN-based method, achieving further performance gains with a lightweight network that trains fast and ensures good performance across different sensors.
We recently proposed a convolutional neural network (CNN) for remote sensing image pansharpening obtaining a significant performance gain over the state of the art. In this paper, we explore a number of architectural and training variations to this baseline, achieving further performance gains with a lightweight network which trains very fast. Leveraging on this latter property, we propose a target-adaptive usage modality which ensures a very good performance also in the presence of a mismatch w.r.t. the training set, and even across different sensors. The proposed method, published online as an off-the-shelf software tool, allows users to perform fast and high-quality CNN-based pansharpening of their own target images on general-purpose hardware.