Machine Learning: How it can help reduce road accidents

Motorcycle crashes account for a disproportionate number of motor vehicle deaths and injuries in the U.S and worldwide. Motorcycle helmet use can lead to an estimated 42% reduction in risk for fatal injuries and a 69% reduction in risk for head injuries. Therefore encouraging the use of helmets can help reduce dramatically the number of fatal injuries.

But how we can encourage the use of helmets and how machine learning can help towards this goal?

InAccel has recently developed SaveDisplay that nudge the use of helmets by motorcycle riders. SafeDisplay can automatically detect of motorcycle riders wear helmets and present this information on a LED display. SafeDisplay can be installed next to the roads similarly to devices that measure the speed and present the speed of the vehicles in their display.

SafeDisplay (patent pending) using a novel technology based on machine learning, computer vision and advanced hardware devices can automatically identify the use of helmets by motorbike riders. The device display the information on the display as a warning in order to nudge the riders to use the helmets.

The main goal of the SafeDisplay device is to encourage the use of the helmets. Specifically, the main goal is to nudge the use of helmets and not to penalize the riders for not using the helmets.

This is why the camera and the display are used only for recognition of the helmet and the motorcycle and not for reading private data such as driving plates.

SafeDisplay has been developed by InAccel, Inc. that has a long experience on machine learning and hardware accelerators. InAccel technology allows the easy utilization of hardware accelerators for applications like machine learning and artificial intelligence. InAccel has been awarded the Venture Impact Award in 2021, supported by The Hellenic Initiative. InAccel is part of Elevate Greece startup ecosystem.

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read More