Smart Traffic Management and Accident Detection
With the help of OpenCV, we are able to calculate crowd density and prevent us from traffic congestion. We are also detecting accidents and providing immediate assisstance without human intervention.
Created on 17th March 2019
•
Smart Traffic Management and Accident Detection
With the help of OpenCV, we are able to calculate crowd density and prevent us from traffic congestion. We are also detecting accidents and providing immediate assisstance without human intervention.
The problem Smart Traffic Management and Accident Detection solves
People would be highly benefitted if our solution comes into play. a lot of time would be saved, reducing traffic jams and also detecting and providing immediate assisstance to accident victims. TensorFlow's Count API and OpenCV helps us to find the intensity of crowd and give more priority to the higher crowd intensity by changing the traffic signals automatically to green and stopping the minimum intensity crowd with red signal.
Our algorithm also detects accidents and other kinds of mishap and takes a screenshot and sends it to the emergency services which may include Police and Medical department. After getting an immediate request, apt services can be provided to the victims.
Challenges we ran into
Deep Learning means hurdles are going to come across. Training a model gets sometimes complicated but we overcame it after trying multiple times and got a good enough model with well accuracy. Taking an image and then uploading it into a website was bit difficult in a single python script. But thanks to Imgur API with their well written documentation which helped us a lot.