Answers are grounded in this paper and cite their sources.
Pansharpening refers to the process of extracting spatial details from a panchromatic (PAN) image and incorporating them into a multi-spectral (MS) image to produce a high spatial resolution MS (HRMS) image. Current pansharpening methods fail to consider the high-frequency spatial details in MS images, leading to spectral distortion in the fused results. To address this issue, the paper proposes a pansharpening method based on adaptive spatial detail fusion and injection weights refinement. First, a multi-stage intensity enhancement module (MIEM) is constructed to enhance the intensity component I of an MS image for spatial-spectral preservation. Second, an adaptive fusion strategy for spatial detail is designed by measuring the correlation and discrepancies between the spatial details of the PAN and I images, ensuring that spatial details from the source images are fused effectively. Third, the jointly weighted injection coefficients are used to refine the fused spatial details based on edge and spectral correlation. Last, the refined spatial details are incorporated into upsampled MS(UPMS) images to produce the final pansharpened images. Both simulated and real experiments have demonstrated that the proposed method outperforms several traditional and deep learning-based approaches. Specifically, the new method enhances the RASE metric by 12.67% and 15.99% relative to the second-best method on the reduced-scale Pléiades and IKONOS datasets, respectively.
No takes yet. Share an insight, caveat, or question.
Dai et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: