/Number_plate_recognition

Primary LanguageJupyter Notebook

Number Plate Recognition using YOLOv5 and Flask

Overview

This project combines YOLOv5, a real-time object detection model, with a Flask web application for number plate recognition. The dataset is annotated using LabelImg, and the app is run using app.py.

Getting Started

Prerequisites

Ensure you have the following installed:

  • Python 3.x
  • PyTorch
  • OpenCV
  • YOLOv5
  • LabelImg
  • Flask

How It Works

YOLOv5

YOLOv5 divides an image into a grid, predicting bounding boxes and class probabilities for objects. Designed for real-time object detection.

LabelImg

LabelImg streamlines manual annotation for object detection. Annotations are saved in YOLO format.

Training

Feed annotated data into YOLOv5. The model adjusts parameters for accurate predictions. Trained weights are saved for later inference.

Inference

The Flask app (app.py) runs inference on new images using the trained YOLOv5 model.