Automatic Number Plate Recognition Using Raspberry Pi

Abstract

Automatic Number Plate Recognition (ANPR) is an important technology used for automatically detecting and reading vehicle number plates using image processing techniques. It plays a major role in modern traffic management systems, parking automation, and security applications. In this project, a low-cost and efficient ANPR system is developed using Raspberry Pi 4 Model B+ along with server integration for data storage and remote access. The system captures vehicle images using a Raspberry Pi Camera Module and processes them using OpenCV. Image preprocessing techniques such as grayscale conversion, edge detection, and contour detection are used to identify the number plate region. After detection, Optical Character Recognition (OCR) is applied using Tesseract to extract the vehicle number. The extracted data is displayed on a TFT display (ST7789) and also sent to a server using HTTP requests. A server-based system is implemented using FastAPI, which stores detected number plates along with timestamps and images. This allows remote monitoring and data management through a web interface. The system works in real-time and provides good accuracy under proper lighting conditions. The main advantage of this system is its low cost, portability, and ability to integrate with cloud or local servers. It is suitable for applications such as parking systems, toll collection, and security surveillance, especially in small-scale environments.