Deep learning startup Neurala Inc. wants to use artificial intelligence for public safety purposes. To that end it’s teaming up with Motorola Solutions Inc. today to develop “intelligent cameras” that can be worn by police officers to help search for missing children and other persons or objects of interest.

Neurala, whose technology was originally designed for NASA to help navigate its Curiosity rover on Mars, has built a deep learning-based “neural network.” It allows network edge devices to learn without the need for servers running in the cloud, enabling a new generation of smart cars, video cameras and industrial machines, according to the company. Neurala recently joined with the Lindbergh Foundation to combine its technology with drones to combat illegal poaching in South Africa, and now it’s hoping to help out with public safety issues at home.

Neurala said it’s working with Motorola to integrate its AI systems into that firm’s cameras, including the Si500 body-worn cameras often worn by law enforcement officers. The idea is to create intelligent cameras that can learn at the network edge in order to identify, search and locate persons or objects of interest. So as soon as a child is reported missing, an image of that child can be sent to all police officers in the area, and their body-worn camera will be able to immediately identify that child if it scans its face, even if an officer misses it.

The companies said this kind of system will significantly reduce the time and effort it takes to locate a missing child in crowded and chaotic environments.

“Neurala’s technology enables AI applications to learn at the edge after their deployment,” Paul Steinberg, chief technology officer at Motorola, said in a statement. “In the case of a missing child, imagine if the parent showed the child’s photo to a nearby police officer on patrol. The officer’s body-worn camera sees the photo, the AI engine ‘learns’ what the child looks like and deploys an engine to the body-worn cameras of nearby officers, quickly creating a team searching for the child.”

Any mention of facial recognition technology usually sends privacy advocates into a panic, but Neurala Chief Executive Massimiliano Versace was quick to play down those concerns. He pointed out that the software doesn’t record any data or images it scans, but instead only looks for a matching face, so privacy violations simply aren’t possible. “We truly believe that the many benefits made possible by this technology, including the ability to more easily find a missing child, will alleviate any misplaced concerns about privacy,” he said.

Video still shot showing missing child located

Versace declined to say if any police forces or other law enforcement agencies had actually expressed an interest in the software at this stage. However, he did say the technology has much larger potential. For example, it could also be used with video cameras at sporting events in order to track the ball in a football game or a specific runner in a racing event.

The intelligent cameras are still some way off in any case. The companies said they’re working to build a prototype that can rapidly apply real-time learning for a person of interest search, but said only that they will provide an update at a later date.

Image: José Morcillo Valenciano/Flickr