UVeye raises $4.5M to use computer vision to inspect underside of vehicles at security checkpoints

UVeye, an Israeli startup that is building computer vision and machine learning technology to be used to help detect security threats by scanning the underside of passing vehicles, has raised $4.5 million in seed funding. The round was led by Ahaka Capital, with participation from angel network SeedIL.

Initially being applied to roadside security — such as stopping car bombs or drugs smuggling — UVeye’s tech claims to be able to analyse any vehicle from underneath to identify and detect threats that would otherwise be concealed to the human eye, even as it is moving, up to 28 MPH, apparently.

It does this using “strategically angled and synchronised hi-res cameras” to build a 360 degree digital model, and says that three seconds after a vehicle passes over UVeye’s ground installed device, the system is able to process multiple images to create a 3D model of the undercarriage and provide high resolution full colour visuals to rule out any security risks.

This is also where UVeye’s combination of vehicle manufacture-supplied data and machine learning kicks in, which can compare and track characteristics of different vehicle models for differentiators, such as weight and part placement. It claims to even be able to recognise a foreign object (or otherwise) the size of a USB stick. The system also uses audio to “listen” for anything unusual.

“UVeye is changing the way people approach security when traveling by vehicle with a fast, accurate and automatic machine learning inspection system that can detect threatening objects or unlawful substances, for example, bombs, unexposed weapons and drugs,” Amir Hever, CEO and co-founder of UVeye, tells me in an email.

“With its uninterrupted traffic flow, UVeye introduces an approach to top security that traditional vehicle inspection methods are missing. We are the first to introduce a machine learning vehicle inspection system that detects anomalies in any vehicle while in motion within three seconds using advanced image processing and audio recordings”.

To date, UVeye systems have been installed within the security market in Israel, Russia, Kenya, Congo, China, and other unnamed countries. Security customers include government offices, embassies, port facilities (sea/air), and critical infrastructures e.g. power plants, gas factories, hotels, and private establishments.

But, noteworthy, the startup is finding a second market for its tech: the wider automotive market. “In the era of Mobility-as-a-Service, companies such as car rental companies, fleets, car dealerships, vehicle repair shops, and OEMs all rely on seamless vehicle operation. UVeye’s machine learning system can detect vehicle leaks, wear and tear, and any damages that would previously go unnoticed. Detecting any of these damages can save companies the time and costs involved in conventional inspection methods,” explains Hever.

To that end, I’m told that UVeye has begun pilot programs with car rental and used car companies. It also plan on expanding its product to fleets, specifically trucks and buses.