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

Urban areas consume over two-thirds of the world's energy and account formore than 70 percent of global CO2 emissions. As stated in IPCC's GlobalWarming of 1.5C report, achieving carbon neutrality by 2050 requires a clearunderstanding of urban geometry. High-quality building footprint generationfrom satellite images can accelerate this predictive process and empowermunicipal decision-making at scale. However, previous Deep Learning-basedapproaches face consequential issues such as scale invariance and defectivefootprints, partly due to ever-present class-wise imbalance. Additionally, mostapproaches require supplemental data such as point cloud data, building heightinformation, and multi-band imagery - which has limited availability and aretedious to produce. In this paper, we propose a modified DeeplabV3+ module witha Dilated Res-Net backbone to generate masks of building footprints fromthree-channel RGB satellite imagery only. Furthermore, we introduce an F-Betameasure in our objective function to help the model account for skewed classdistributions and prevent false-positive footprints. In addition to F-Beta, weincorporate an exponentially weighted boundary loss and use a cross-datasettraining strategy to further increase the quality of predictions. As a result,we achieve state-of-the-art performances across three public benchmarks anddemonstrate that our RGB-only method produces higher quality visual results andis agnostic to the scale, resolution, and urban density of satellite imagery.

A Semantic Segmentation Network for Urban-Scale Building Footprint Extraction Using RGB Satellite Imagery