ICVLPR

ICVLPR
· Source

ICVLPR (Indonesian Commercial Vehicle License Plate Recognition) is a deep learning project that uses a modified LPRNet architecture to recognize license plates, specifically tailored for commercial vehicles in Indonesia. It was developed as my final undergraduate thesis project at Universitas Diponegoro.

Features

  • End-to-end OCR for license plates, using CNN + CTC decoder
  • Trained on a custom dataset of Indonesian commercial vehicles
  • Lightweight and inference-ready model
  • Preprocessing pipeline: resizing and image normalization
  • Evaluation tools and character-level accuracy metrics

Use Cases

This project can be used for:

  • Automated toll gate systems
  • Vehicle fleet monitoring
  • Logistics and transport automation

Want to try it or contribute? Check out the source code on GitHub.