Obstacles detection (cars, pedestrians, …) by Computer Vision and Deep Learning



. Detects obstacles, their nature (car, pedestrian, …) and location. ADAS and Autonomous Vehicle.


. Computes interdistance (between car and obstacle) and sends it to SafetyNex. Short interdistance is a factor of risk.




software component



Road(t) :
Coordinates of road borders
Coordinates of free space borders (even if there are no markings)
In the image space in pixels In the real 3D world in meters



NEXYAD algorithm ObstaNex


Detect cars
Detect vehicles
Simply a must have. But runs in real time on a regular smartphone (for instance) instead of :
. A special chip
. A special heavy computing architecture
It changes a lot the price of mass volume deployment.