Earth Observation and Machine Learning to Meet Sustainable Development Goal 8.7: Mapping Sites Associated with Slavery from Space

Abstract

A large proportion of the workforce in the brick kilns of the Brick Belt of Asia are modern-day slaves. Work to liberate slaves and contribute to UN Sustainable Development Goal 8.7 would benefit from maps showing the location of brick kilns. Previous work has shown that brick kilns can be accurately identified and located visually from fine spatial resolution remote-sensing images. Furthermore, via crowdsourcing, it would be possible to map very large areas. However, concerns over the ability to maintain a motivated crowd to allow accurate mapping over time together with the development of advanced machine learning methods suggest considerable potential for rapid, accurate and repeatable automated mapping of brick kilns. This potential is explored here using fine spatial resolution images of a region of Rajasthan, India. A contemporary deep-learning classifier founded on region-based convolution neural networks (R-CNN), the Faster R-CNN, was trained to classify brick kilns. This approach mapped all of the brick kilns within the study area correctly, with a producer’s accuracy of 100%, but at the cost of substantial over-estimation of kiln numbers. Applying a second classifier to the outputs substantially reduced the over-estimation. This second classifier could be visual classification, which, as it focused on a relatively small number of sites, should be feasible to acquire, or an additional automated classifier. The result of applying a CNN classifier to the outputs of the original classification was a map with an overall accuracy of 94.94% with both low omission and commission error that should help direct anti-slavery activity on the ground. These results indicate that contemporary Earth observation resources and machine learning methods may be successfully applied to help address slavery from space.

Impact of Urban Growth on Air Quality in Indian Cities Using Hierarchical Bayesian Approach

“…Official estimate of the number of brick kilns in India is unavailable and they are assumed to be more than 100,000 [34]. Brick kilns are often found around the periphery of urbanizing cities and can be spotted in high resolution imagery due to their distinct appearance [61]. They were visually identified around the study sites using Google Earth imagery.…”

Deep Learning Based Fossil-Fuel Power Plant Monitoring in High Resolution Remote Sensing Images: A Comparative Study

“…For example, He et al [1], Feng et al [2] and Kai et al [3] analyzed the haze pollution using satellite remote sensing data. Recently, Foody et al [38] trained Faster R-CNN to localize and classify brick kilns in RSIs in order to contribute to Sustainable Development Goal. Yao et al [4] presented the most related previous work followed by us.…”

Digital Earth for Sustainable Development Goals

“…Traditional EO techniques are designed for extracting information on environmental variables, with only a few being related to the built environment and associated infrastructure (e.g., built-up areas and roads). Hence, there are limitations on the possibility of EO alone producing reliable metrics for SDG indicators (see Table 13.1); however, approaches underpinned by big Earth data do have some potential, as evidenced in recent research by Kussul et al (2019), Foody et al (2019), Freire et al (2018), and Corbane et al (2017). Specifically:…”

“…• meta-optimization of EO with external data-intensive infrastructure has led to improved mapping of built-up areas in support of the global human settlement layer (Corbane et al 2017) • national mapping of SDG indicators 15.1.1, 15.3.1 and 2.4.1 has been achieved through synergy of in situ and multi-resolution satellite data (Kussul et al 2019) • big Earth Data (global census data and satellite-derived built-up area maps) has enabled enhanced mapping of population distribution along coastlines (Freire et al 2018) • EO and machine learning have enabled mapping of sites associated with slavery, in support of SDG target 8.7 ("take immediate and effective measures to eradicate forced labour, end modern slavery and human trafficking and secure the prohibition and elimination of the worst forms of child labour") (Foody et al 2019). In summary, EO data does not directly deliver the SDG indicators agreed by the Inter-Agency and Expert Group (IAEG) on SDGs; rather, it provides a diversity of spatio-temporal information that can then be related to the indicator framework.…”