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 integration of multisource remote sensing data and deep learning modelsoffers new possibilities for accurately mapping high spatial resolution forestheight. We found that GEDI relative heights (RH) metrics exhibited strongcorrelation with the mean of the top 10 highest trees (dominant height)measured in situ at the corresponding footprint locations. Consequently, weproposed a novel deep learning framework termed the multi-modal attentionremote sensing network (MARSNet) to estimate forest dominant height byextrapolating dominant height derived from GEDI, using Setinel-1 data, ALOS-2PALSAR-2 data, Sentinel-2 optical data and ancillary data. MARSNet comprisesseparate encoders for each remote sensing data modality to extract multi-scalefeatures, and a shared decoder to fuse the features and estimate height. Usingindividual encoders for each remote sensing imagery avoids interference acrossmodalities and extracts distinct representations. To focus on the efficaciousinformation from each dataset, we reduced the prevalent spatial and bandredundancies in each remote sensing data by incorporating the extended spatialand band reconstruction convolution modules in the encoders. MARSNet achievedcommendable performance in estimating dominant height, with an R2 of 0.62 andRMSE of 2.82 m, outperforming the widely used random forest approach whichattained an R2 of 0.55 and RMSE of 3.05 m. Finally, we applied the trainedMARSNet model to generate wall-to-wall maps at 10 m resolution for Jilin,China. Through independent validation using field measurements, MARSNetdemonstrated an R2 of 0.58 and RMSE of 3.76 m, compared to 0.41 and 4.37 m forthe random forest baseline. Our research demonstrates the effectiveness of amultimodal deep learning approach fusing GEDI with SAR and passive opticalimagery for enhancing the accuracy of high resolution dominant heightestimation.

Multimodal deep learning for mapping forest dominant height by fusing GEDI with earth observation data