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

As a unique classification scheme for urban forms and functions, the localclimate zone (LCZ) system provides essential general information for anystudies related to urban environments, especially on a large scale. Remotesensing data-based classification approaches are the key to large-scale mappingand monitoring of LCZs. The potential of deep learning-based approaches is notyet fully explored, even though advanced convolutional neural networks (CNNs)continue to push the frontiers for various computer vision tasks. One reason isthat published studies are based on different datasets, usually at a regionalscale, which makes it impossible to fairly and consistently compare thepotential of different CNNs for real-world scenarios. This study is based onthe big So2Sat LCZ42 benchmark dataset dedicated to LCZ classification. Usingthis dataset, we studied a range of CNNs of varying sizes. In addition, weproposed a CNN to classify LCZs from Sentinel-2 images, Sen2LCZ-Net. Using thisbase network, we propose fusing multi-level features using the extendedSen2LCZ-Net-MF. With this proposed simple network architecture and the highlycompetitive benchmark dataset, we obtain results that are better than thoseobtained by the state-of-the-art CNNs, while requiring less computation withfewer layers and parameters. Large-scale LCZ classification examples ofcompletely unseen areas are presented, demonstrating the potential of ourproposed Sen2LCZ-Net-MF as well as the So2Sat LCZ42 dataset. We alsointensively investigated the influence of network depth and width and theeffectiveness of the design choices made for Sen2LCZ-Net-MF. Our work willprovide important baselines for future CNN-based algorithm developments forboth LCZ classification and other urban land cover land use classification.

Multi-level Feature Fusion-based CNN for Local Climate Zone Classification from Sentinel-2 Images: Benchmark Results on the So2Sat LCZ42 Dataset