English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

The process of fusing a high spatial resolution (HR) panchromatic (PAN) imageand a low spatial resolution (LR) multispectral (MS) image to obtain an HRMSimage is known as pansharpening. With the development of convolutional neuralnetworks, the performance of pansharpening methods has been improved, however,the blurry effects and the spectral distortion still exist in their fusionresults due to the insufficiency in details learning and the frequency mismatchbetween MSand PAN. Therefore, the improvement of spatial details at the premiseof reducing spectral distortion is still a challenge. In this paper, we proposea frequency-aware fusion network (FAFNet) together with a novel high-frequencyfeature similarity loss to address above mentioned problems. FAFNet is mainlycomposed of two kinds of blocks, where the frequency aware blocks aim toextract features in the frequency domain with the help of discrete wavelettransform (DWT) layers, and the frequency fusion blocks reconstruct andtransform the features from frequency domain to spatial domain with theassistance of inverse DWT (IDWT) layers. Finally, the fusion results areobtained through a convolutional block. In order to learn the correspondence,we also propose a high-frequency feature similarity loss to constrain the HFfeatures derived from PAN and MS branches, so that HF features of PAN canreasonably be used to supplement that of MS. Experimental results on threedatasets at both reduced- and full-resolution demonstrate the superiority ofthe proposed method compared with several state-of-the-art pansharpeningmodels.

Pansharpening via Frequency-Aware Fusion Network With Explicit Similarity Constraints