Mass Image Labeling via Citizen Science for Deforestation Detection

• Hugo Resende UNIFESP
• Álvaro L. Fazenda UNIFESP
• Fabio A. Faria UNIFESP / Universidade de Lisboa

DOI: https://doi.org/10.5753/wcama.2025.7514

Abstract

Citizen science is an essential tool for democratizing knowledge, as it allows anyone to assist in large-scale data analysis. In the environmental context, this approach is essential for monitoring threats such as tropical forest deforestation. In this regard, this study demonstrates how massive volunteer participation can generate reliable labels through the wisdom of crowds. In the ForestEyes project, 1,800 remote sensing images were collaboratively classified and labeled (with an average of 875 classifications per day), achieving 88.78% accuracy compared to reference data, highlighting the potential of citizen science as a support tool for deforestation detection.

References

Achanta, Radhakrishna et al. (2012). Slic superpixels compared to state-of-the-art superpixel methods. IEEE TPAMI, 34(11):2274–2282.

Artaxo, Paulo wt al. (2022). Tropical and boreal forest atmosphere interactions: A review. Tellus. Series B: Chemical and Physical Meteorology, 74:24–163.

Brown, K. and Pearce, D. W. (2023). The causes of tropical deforestation: the economic and statistical analysis of factors giving rise to the loss of the tropical forests. Taylor & Francis.

Dallaqua, F. B., Fazenda, Á. L., and Faria, F. A. (2021). Foresteyes project: Conception, enhancements, and challenges. Elsevier FGCS, 124:422–435.

Diniz, Cesar G. et al. (2015). Deter-b: The new amazon near real-time deforestation detection system. IEEE J-STARS, 8(7):3619–3628.

Drusch, M. et al. (2012). Sentinel-2: Esa’s optical high-resolution mission for gmes operational services. Remote Sensing of Environment, 120:25–36.

Fazenda, A. L. and Faria, F. A. (2024). Foresteyes: Citizen scientists and machine learning-assisting rainforest conservation. Communications of the ACM, 67(8):95–96.

INPE (2024). PRODES - Project for Monitoring Deforestation in the Legal Amazon by Satellite. [link]. Acessado em setembro/2024.

Lapola, David et al. (2023). The drivers and impacts of amazon forest degradation. Science, 379(6630):eabp8622.

Mach, E. (1865). Über die Wirkung der räumlichen Vertheilung des Lichtreizes auf die Netzhaut. K. k. Hof-und Staatsdruckerei.

Main-Knorn, M. et al. (2017). Sen2Cor for Sentinel-2. In Image and Signal Processing for Remote Sensing, volume 10427, page 1042704.

Pro, G. F. W., Watcher, F., and Atlases, F. (2023). Global forest watch. Update.

Reis, E. and Guzman, R. (2023). An econometric model of amazon deforestation. In The causes of tropical deforestation, pages 172–191. Routledge.

Resende, Hugo et al. (2024). Sampling strategies based on wisdom of crowds for amazon deforestation detection. In SIBGRAPI, pages 1–6. IEEE.

Roy, David P. et al. (2014). Landsat-8: Science and product vision for terrestrial global change research. Remote sensing of Environment, 145:154–172.

Simpson, Robert et al. (2014). Zooniverse: observing the world’s largest citizen science platform. In ACM WWW, pages 1049–1054.