TraffiSense assists drivers in driving carefully by keeping an eye on their speed and notifying them of overspeeding. Thereby, eliminating the chance of accidents.
We initially planned on using OpenCV to detect the traffic but later on realised that HAAR Cascades isn't the proper way to do it since there's a lot of overlapping objects in traffic. So we ultimately had to implement a DL model using YOLOv3 to detect the pedestrians and vehicles. This was used to calculate the traffic density which was used by the backend to calculate a top speed parameter that was sent to the vehicles in the vicinity.
Discussion