The “miracle” backpack for the blind

AI technology in a backpack for the blind

Autonomous vehicles and robots use complex sensors and cameras to perceive their surroundings – but people with visual impairments can still use canes and guide dogs. Now, engineers have developed a wearable, voice-activated system that can track obstacles in real time, and describe the surroundings to a visually impaired person.

This new vision aid system is made up of several components that can be worn without being too bulky. It just includes the vest, a backpack and a pair of headphones. The team says that hiding electronic devices is a key goal.

In this system, the shirt will integrate a hidden 4K camera that provides color information and a pair of stereoscopic cameras that map depth of field. This visual information is then fed to OpenCV’s Artificial Intelligence with Depth Kit (OAK-D) which is stored in a backpack.

OpenCV’s Artificial Intelligence with Depth (OAK-D) suite is tasked with using neural networks to analyze image data. It also contains a portable battery that provides up to eight hours of use and works with a USB-connected GPS unit.

The resulting image data is then relayed via Bluetooth to a pair of headphones, allowing the user to know what’s around them. It can warn of obstacles of different shapes, sizes and types, and declare their position in relation to location to visually impaired users.

For example, the system could notify someone walking down the street that they are approaching a trash can at the “bottom left” or a low branch with “top, center”. Stumbling hazards like curbs or stairs can be detected during elevation changes, and the system can even recognize important things like Stop signs.