This project was selected out of the five problem statements in round 2 of TCS-HumAIn
The project is developed using TensorFlow to detect an Indian License Plate from a car and to recognize the charactes from the detected plate.
Approach:
1)Detect License Plate using CCA (Connected Component Analysis) 2)Perform segmentation of characters (applying CCA on the plate) 3)Train a CNN model using images of characters (numbers and alphabets) 4)Prediction of characters in License Plate
Run main.py
Tech-stack:
Python 3.7, tensorflow, keras, open-cv
Software packages:
Anaconda 3 (Tool comes with most of the required python packages along with python3 & spyder IDE)