E-Challan-License Plate Recognition

Project Overview

• Training a machine learning model to detect and recognize license plates.
• Creating an electronic challan interface that uses the trained models.
• Use the license plate extracted from the model to fetch challan details.
• Visualizing all the data about challans based on different details.

Code and Resources Used

OS: Windows_NT x64 10.0.19045
Microsoft Visual Studio: 1.90.0 (user setup)
Python: 3.12
MongoDB: 8.0 Express.js: 4.19.2
React.jS: 18.2.0
Node.js: 20.9.0
Tremor: 3.16.3
Axios: 1.7.2
Opencv Python: 4.9.0.80
Ultralytics: 8.2.18
Google Cloud Storage: 1.25.0
EasyOCR: 1.7.1
Pillow: 10.3.0

Activity Diagrams

HomePage

Post-Image Upload & Fetch

Visualization

Machine Learning Model

Model Used: YOLO V8 (Ultralytics)
GPU Used: T4 GPU (Google Colab)
Data Anottation Platform: Roboflow
Dataset: Indian Vehicle Dataset
Deployment: Google Cloud Function APIs

Model Stats

F1-Confidence Curve

Precision Confidence Curve

Precission-Recall Curve

Confusion Matrix

Multiple Results

Training Batch

Validation Batch

More stats are available in the main/Stats

Useful References

YOLO (You Only Look Once) Object Detection Model: YOLO Official Website
EasyOCR Library: EasyOCR GitHub Repository
Google Cloud Functions Documentation: Google Cloud Functions
MERN Stack Guide: MERN Stack Guide
LDRS Icons :LDRS Github
MongoDB Setup, Fetching: Video Guide #1 , Video Guide #2
React JS & API Guide :Video Guide
Tremor Guide : Video Guide
Node JS & MongoDB Connect: Video Guide