Smart vehicle camera automatically detects IED

0
43

Improvised Explosive Devices(IED) or roadside bombs are tricky and effective killers. They are easy to make and hide, making it the weapon of choice for terrorists and insurgents across the world. Unearthing and disabling these lethal devices is hard. A new vehicle-mounted system, though, is designed to “spot the signs” of IEDs.

Electrical engineer, Dennis van de Wouw of Eindhoven University of Technology (TU/e), has, in close association with experts and industry of the Netherlands Ministry of Defense, made a real-time early-warning system.

When attached to a military vehicle, it can automatically identify the presence of those bombs by registering doubtful changes in the environment.

Improvised Explosive Devices (IED) are a severe danger to both civilians and troops in conflict zones. They are easily invented using everyday household items, and, with instructions extensively available online, their construction no more requires military training. Due to their high effect (more than 66% of coalition fatalities in the war in Afghanistan are a result of IEDs), ease of construction, and difficulty to detect, they have become the favorite weapon for terrorists and insurgents in war zones across the world.

In order to reduce losses during military transports, recurrent surveillance of the high-risk routes is obligatory. One useful method of inspection is ground-based patrols, which confine potential threats by probing for suspicious patterns in the environment. This is a very challenging task, because humans have difficulties engaging with a task for a longer time in a strange environment. Thus, an urgent need arises for a system that can help military personnel in finding possible threats during driving.

ANY SIZE, SHAPE OR COLOR

Dennis van de Wouw and his project partners built a real-time early-warning system for IEDs that is mounted on a vehicle. This system is competent to identify unknown objects and doubtful changes in the environment automatically. Automated detection of such mysterious objects is challenging using traditional detection techniques, as it is simply not recognized what the IED looks like. It could have any shape, size and color.

In its place, they started from the guess it that placing an IED causes slight changes to the environment, such as plowing tracks or newly emerged objects, like triggers used by terrorists or insurgents to explode the IED. Consequently, Van de Wouw armed a military vehicle with an intelligent video system that is able to automatically find doubtful changes in the environment while driving. This system consists of a GPS positioning system, a stereo camera which is a camera with two lenses that produce 3D-images, and an image analysis platform.

IED detection

The system mechanically records the environment and matches it to images recorded earlier, for example during an earlier clearance operation. The system then examines all changes. Only those changes that specify the possible presence of an IED are shown to the military operator through an interactive Graphical User Interface. “This enables a military operator to take suitable actions, such as stopping in time for more inspection of the potential threat, or avoiding it altogether,” says Van de Wouw.

Challenge
His main challenge was making a system that was able to match accurately historical and live images of the environment. This is not insignificant: the view of the recordings may differ, as well as the time of day or weather conditions, creating confusing changes in lightning and perspective. Using stereo cameras, automated image analysis and advanced 3D modelling, Van de Wouw was able to design a robust system that is able to detect suspicious changes in the environment effectively. Furthermore, he managed first exploratory experiments with Artificial Intelligence, which show favorable detection capabilities even under harsh lighting conditions.

The developed system has been successfully proven to the Dutch Ministry of Defense, showing its detection capabilities on a military landscape. There, it was able to identify all ‘unknown’ test objects and changes with at least medium contrast. The Ministry is consequently pleased with the work of Van de Wouw and his associates, especially with the speed at which the new system handles the warning signals.

“The danger warning has to come in time so that the military vehicle can stop at a cautious distance from the roadside bomb. This puts great pressure on the time-management for parallel processing. So far all the results look very hopeful”, says Silvester de Bruin, innovation adviser at the Ministry.

According to Van de Wouw, the new early-warning system is one of the few operational change detection systems for counteracting IED capable of detecting small suspicious changes in the environment in real-time. He is nowadays working for ViNotion, a spin-off of the Video Coding and Architectures research group at TU/e, where he is developing this technology in close association with the Ministry of Defense.

 

Leave a reply