Doreen Boyd remembers the first time she saw a hint of slavery from space. A 2017 satellite image of the state of Rajasthan in India showed a brown oval that looked like a dusty high school track. But nothing was so trivial: she knew it was a brick kiln, one of tens of thousands in South Asia that is often subjected to forced labor. Boyd, director of the data program at the Rights Lab of the University of Nottingham in the UK, realized that such images could help him to account for ovens, allowing organizations on the field to target the slave owners on the sites. "You can not see slavery directly, but you can deduce it," she says.
The increase in the number of Earth observation satellites, as well as the improved algorithms to interpret the deluge of data that they provide, highlight the importance of the Earth's observation satellites. modern slavery. This week, at a conference in New York sponsored by the United Nations University (UNU), computer scientists, slavery experts and political strategists presented the latest efforts in their field and ways to work together in the middle of the brain. "We're doing science as a team," said Austin Choi-Fitzpatrick, peace studies expert at the University of San Diego, California, who interviewed slave owners at firing sites like Space Lab.
Some 40.3 million people are held in slavery today, according to the latest estimates from the International Labor Organization, headquartered in Geneva, Switzerland. But finding them is difficult. "Those affected are often hidden from the state," said James Cockayne, director of the UNU Center for Policy Research in New York, who helped organize the conference. Boyd estimates, however, that one-third of slavery is visible from the space, whether it is scars from ovens or illegal mines or contours of food processing camps. fish in transit.
In 2015, DigitalGlobe, whose Earth Observation satellites provide much of the data for Google Earth, recruited users to zoom in on images of Lake Volta, Ghana, where experts suspect children of To be forced to work. "We were looking for boats across the huge lake," says Rhiannan Price, director of DigitalGlobe's global development program in Westminster, Colorado. In total, 90,000 users blocked 80,000 boats, buildings and fish cages.
Boyd now uses artificial intelligence to speed up searches. As a pilot project, she and her colleagues at the Rights Lab used visual researchers using participatory sources to identify brick kilns. The oval shape of the big ovens, sometimes 150 meters long, and their chimneys are distinctive even from the space. "You can not mix them with anything else," says Boyd.
Since then, Boyd has turned to machine learning algorithms that recognize dryers after being trained in human-marked examples. Last month, in Remote Sensing, she and her colleagues said that the algorithms could correctly identify 169 of the 178 driers in Google Earth in a region of Rajasthan, although nine false positives were also generated.
Planet, another company, has about 150 small satellites that capture images of the Earth's entire land mass every day. Images have a lower resolution than DigitalGlobe, but their frequency opens up opportunities to identify changes over time. "Every day we see all the buildings, all the fields, all the mines, all the quarries, all the forests," says Andrew Zolli, vice president of Planet, responsible for global impact initiatives in New York.
With Planet Data, Boyd and the Rights Lab plan to investigate the rapid signatures of slavery. From the space, you can watch a cotton crop in Turkmenistan and, depending on how quickly the cotton disappears, you can find out if the machines or hands have harvested it. In the Sundarbans, an area encompassing India and Bangladesh, shrimp farms and fish processing camps employ slaves to clear mangrove trees – a process that satellites can capture.
Rights Lab also plans to use satellite data in other parts of the spectrum. The European Space Agency's Sentinel-1 satellite uses a radar to measure tiny variations in altitude, which could reveal the ground subsidence of a mining tunnel with clandestine workers inside. The agency's Sentinel-2 satellite, which detects frequencies of infrared light, can highlight mining operations based on the reflections of recently exposed minerals.
The sky-high solutions, even the smart ones, are not a panacea. Other organizations need to convert space detective work into action on the ground. And even before the start of the analysis, researchers must know what they are looking for. "The images themselves are significantly less profitable if you do not have local knowledge of what you are looking at," says Zolli.