Smart Traffic Control System

This was the AI code that was designed for Traffic Management purposes in Rakuten's Rakathon 2.0

Was meant to be implemented on a Raspberry Pi with a camera module.

The Inspiration and the backstory is documented in Rakathon2.pdf

Features

• Detects number of different vehicles in the scene and gauges the time required for each different vehicles.

• Prioritizes pedestrian safety.

• Designed to prioritize Ambulance passage.

Requirements:

• OpenCV- pip install opencv-python
• Numpy- pip install numpy
• An appropriate dataset like coco.

How to use

• Install the requirements.
  1. Place the image files in the /images directory.
  2. Choose the image which you would like to analyze and copy its directory in your clipboard.
  3. Use the argument -i to input the path of the image file and -y to input the path of the YOLO model in the command line to run the code.

Known Errors

Incorrect labelling of Automobiles.

Solution: Train the model with the right class-labelling.