AI Based Traffic Control System

Our project is about detection of the number plates of the vehicles at the toll plaza. One time when I was travelling with my family in a car, I'm super excited cause it's my long awited trip that is hapenning finally. While we were on the highway we came across a toll gate where we waited about an hour just to pass through it. I thought why don't we replace fastTags with something that are very fast and secure. Then for the AMD hackathon my friend suggested this idea that we can scan the vehicles's number plate or using RF id we can directly pay the toll gate charges. So I joined in his team. Our idea is to develop a system or model that can scan the number plates and automatically the money will be deducted. And we are selected for the receiving of the kit, Kria KR260 Robotics Starter kit, which is very useful for high performance interfaces Here we have to train the model by using machine learning and deep learning that the data is stored in the database of the system. The editor can view the data that is stored at anytime. Here we used python source code to train the model. The camera will detect the vehicle's number plate and gives us the information. The major disadvantage is that as for now we can't link it to any UPI cause it requires more authentication and permissions from many sources and it'd be a risk of using them now. Other than linking the UPI we succesfully detected the vehicle using the model. Our main components are Kria KR260 Robotics Starter kit and camera from Zebronics. We have connected the camera to the KR260, which gives us the images of the detection.