Detecting global urban expansion over the last three decades using a fully convolutional network

TLDR

The effective detection of global urban expansion is essential for understanding urban sustainability. The study proposes a fully convolutional network to detect global urban expansion from 1992 to 2016. The FCN integrates multi‑source remote‑sensing data, fuses multi‑scale features, and mitigates limited historical training samples to improve accuracy. The FCN identified a 1.3‑fold increase in global urban land area from 274.7 k km² in 1992 to 621.1 k km² in 2016, achieving an average kappa of 0.5—0.3 higher than existing datasets and demonstrating strong potential for effective global urban expansion detection.

Abstract

The effective detection of global urban expansion is the basis of understanding urban sustainability. We propose a fully convolutional network (FCN) and employ it to detect global urban expansion from 1992–2016. We found that the global urban land area increased from 274.7 thousand km2–621.1 thousand km2, which is an increase of 346.4 thousand km2 and a growth by 1.3 times. The results display a relatively high accuracy with an average kappa index of 0.5, which is 0.3 higher than those of existing global urban expansion datasets. Three major advantages of the proposed FCN contribute to the improved accuracy, including the integration of multi-source remotely sensed data, the combination of features at multiple scales, and the ability to address the lack of training samples for historical urban land. Thus, the proposed FCN has great potential to effectively detect global urban expansion.