Skip to content

License plate recognition - Computer Vision

Published: at 03:57 PM

Using computer vision to automatically find and identify a license plate from an input image of a car.

Table of contents

Open Table of contents

Dataset

The environment defined is a 4x4 grid where the agent has to reach the goal state from the initial state, the environment defined is identical for both deterministic and stochastic environments.

The dataset consists of images taken from the Google Open Images Dataset, the Open Images Dataset consists of millions of images that can be used to train and test AI models. For this project images of vehicles are collected.

Google drive link: https://drive.google.com/drive/folders/1-KHew3PpOQpHAuA0g_RA91CgdlN50hZ3?usp=share_link

Pre-processing of the dataset:

Algorithm

The model is trained on the yolov7 algorithm using transfer learning and weights from the yolov7 model Yolov7:

Aspects of the algorithm I’ve coded on my own:

Screenshots

Read more - Project Report