New framework enhances remote sensing image fusion with frequency-independent feature learning

A research team led by Prof. XIE Chengjun and ZHANG Jie from the Hefei Institutes of Physical Science of the Chinese Academy of Sciences, developed a frequency domain-independent feature learning framework. It allows for better representation and fusion of different types of remote sensing images. 

This work was recently published in IEEE Transactions on Circuits and Systems for Video Technology (IEEE TCSVT).

Pan-sharpening, a critical technology in remote sensing image processing, combines high-resolution panchromatic images with low-resolution multispectral images to produce detailed high-resolution multispectral images. This technology is vital for enhancing the balance between spatial and spectral resolution in optical remote sensing satellites. However, current pan-sharpening methods often falter when dealing with out-of-distribution data, as they assume identical data distributions in training and testing datasets.

To address these challenges, the team introduced a frequency decoupled domain-independent feature learning framework. This approach analyzes domain-independent information distribution in image amplitude and phase components, utilizing frequency information separation modules and learnable high-frequency filters to decouple image information. The processed information goes through two dedicated sub-networks, and the final step adjusts feature channels dynamically to improve image fusion and quality.

Cross-scenario tests on multiple public datasets demonstrate the framework's strong generalization performance, effectively handling diverse data distributions. Training on the WorldView-III dataset, the method was tested on other datasets, maintaining excellent performance on the training set and outperforming other methods on generalization datasets. 

Visual comparisons confirm that this framework effectively extracts and learns information, ensuring consistent performance even with varying data distributions.

This advancement marks a significant step forward for applications requiring high-fidelity image data across a wide array of satellite imaging scenarios, according to the team.