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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The effective combination of the complementary information provided by thehuge amount of unlabeled multi-sensor data (e.g., Synthetic Aperture Radar(SAR) and optical images) is a critical topic in remote sensing. Recently,contrastive learning methods have reached remarkable success in obtainingmeaningful feature representations from multi-view data. However, these methodsonly focus on image-level features, which may not satisfy the requirement fordense prediction tasks such as land-cover mapping. In this work, we propose aself-supervised framework for SAR-optical data fusion and land-cover mappingtasks. SAR and optical images are fused by using multi-view contrastive loss atimage-level and super-pixel level in the early, intermediate and later fashionindividually. For the land-cover mapping task, we assign each pixel aland-cover class by the joint use of pre-trained features and spectralinformation of the image itself. Experimental results show that the proposedapproach achieves a comparable accuracy and that reduces the dimension offeatures with respect to the image-level contrastive learning method. Amongthree fusion fashions, the intermediate fusion strategy achieves the bestperformance. The combination of the pixel-level fusion approach and spectralindices leads to further improvements on the land-cover mapping task withrespect to the image-level fusion approach, especially with few pseudo labels.

Self-supervised SAR-optical Data Fusion and Land-cover Mapping using Sentinel-1/-2 Images