DigitalGlobe Spotlights Human Rights Violations with Satellite Imagery and Geoint

In July 2018, a deeply disturbing and violent video began to circulate on social media. Taking place in Cameroon, it depicts two women and two young children being led at gunpoint away from a village by a group of Cameroonian soldiers. Blindfolded, the victims are forced to the ground and shot 22 times by the soldiers.

Investigation by Aliaume Leroy and Ben Strick

Produced by Daniel Adamson and Aliaume Leroy

Motion Graphics: Tom Flannery

The government of Cameroon initially dismissed the video as “fake news.” But BBC Africa Eye, an investigative branch of the BBC that focuses on investigative journalism in Africa, did a thorough investigation through forensic analysis of the footage. Using satellite imagery as well as matching of landmarks and even the types of uniforms worn by the soldiers, they can prove exactly where this happened, when it happened, and who is responsible for the killings.

In a conversation with Steve Wood, senior analyst for DigitalGlobe and their Maxar News Bureau, he talked about his involvement in bringing open source and geoint to the telling of this very shocking story.

“We can now expose things that were previously difficult to get exposed,” said Wood. Wood has been with DigitalGlobe for quite awhile as a career imagery analyst and has a lot of time and experience working with satellite imagery both in the government and at DigitalGlobe.

“Part of what we have been doing here at DigitalGlobe and at Maxar, our parent company, is we have a team called the news bureau and we work very closely with media organizations around the world to provide our imagery and help them tell stories like this,” Wood said.  We like to specialize in the more investigative aspect, although we often get caught up in the current events, whether its Hurricane Florence or a current activity that can be more visually presented using satellite imagery. We’ve worked with the BBC for quite a while on stories in the middle east, Africa and elsewhere.”

Wood received an inquiry that came in through the team, looking for satellite imagery that was related to this Cameroonian video. “Ultimately, I got in touch with the analyst who was producing most of this, Aliaume Leroy. This analyst has been doing work like this for Bellingcat, an investigative organization that with its investigative partner The Insider – Russia have established conclusively the identity of one of the suspects in the poisoning of Sergey and Yulia Skripal, and in the homicide of British citizen Dawn Sturgess using open source geospatial intelligence.”

Bellingcat pulls together open source data, Google Earth, satellite imagery and geospatial activity and pieces it together into a compelling geospatial profile.

Leroy and investigator Ben Strick have done some of this type of work in the past, such as working in the Middle East, and on the story of the Malaysian airline crash.

“They reached out to us to see if we could get any satellite imagery of this village that they believed would likely be the location that was first seen on social media,” said Wood. “They provided me with coordinates and locations, and I went into our archives and looked at imagery that correlates to the timeframe based on the video. I was able to find a number of images and began the dialog with the BBC about this. Ultimately, we collaborated –  I found imagery on this, traded emails, videos, and sent images back and forth, collaborated on identifying some of the features that we believed could be seen in the video and then correlated them with satellite imagery.”

The team at the BBC felt that they had the right area but then were challenged to figure out the timeframe. Correlating what was seen in DigitalGlobe imagery with the rough time frame of the scene in the video was challenging.

“We collaborated, compared thoughts and ideas such as, did you see that tree in this video, the building’s roof was gone in this time frame,” said Wood. “We started narrowing down based on a set of images what appeared to be the right time frame for when the video was taken. We ultimately provided imagery to them which they were then able to incorporate and put into the video itself.”

The BBC team figured out the time of the season and dialed it into the actual days themselves, by comparing the lighting in the imagery and video.

The video of the execution that was being shown on social media was of grainy quality, so it was difficult to see.

“But in that video, we were able to pick out a number of buildings that were on the periphery as the soldiers are walking down the road leading the women and children to their execution,” Wood pointed out. “In that video we really focused on those buildings, because we noticed on the satellite imagery, on several of the image dates, some of the buildings have rooves and on other dates, they did not. So, we were able to take that time frame of the imagery that had the roof and compare to those that didn’t, to then dial in again and get tighter parameters. Then we knew when the video must have been actually taken. It was a good example of taking information that was derived from video and using the satellite imagery as a way to fact check and to correlate that provided additional information to determine when and where the execution occurred.”

Africa Eye mined social media to pull the information about the weapons and the soldiers’ uniforms using open source.

“This story was one of the better ones in my career, as far as being able to follow the trail and piece it together,” Wood said. “One of the things we’re proud about is when we can see our imagery go to benefit a story like this. It appears several of those Cameroon soldiers are awaiting trial and it sounds like finally the government has taken responsibility for it. It’s rewarding to see how geoint can be brought together in this type of effort. It was obviously a tragic and terrible event, but geoint can be used to complete the story.”

DigitalGlobe images around the world every day. Part of their business construct is to have a constantly refreshing map of the earth. Whether the imagery is for immediate use or use later on, it is available as the company has a rich repository dating back to almost 1999.

“We just happened to have images of that village at the right time and the right place,” said Wood. “We were able to pinpoint the month and year. We’ve used our satellite imagery quite a bit in the last couple of years for stories of persecution such as in Myanmar for human rights groups such as Amnesty International, to show the plight of people. International organizations and open source watchdog agencies have been using geospatial information and imagery to expose these human rights violations. It is one of the real benefits to having this technology out there.”

Tags: cloud, crowdsourcing, data, DigitalGlobe, geospatial, GIS, Google, hurricanes, intelligence, mapping, remote sensing, satellite imagery, social media

Categories: analytics, Big Data, cloud, data, DigitalGlobe, geospatial, GIS, Google, government, location based services, location intelligence, mapping, Open Source, photogrammetry, remote sensing, satellite imagery, spatial data

This entry was posted on Friday, September 28th, 2018 at 1:30 pm. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.