Efficient Video-Based ALPR System Using YOLO and Visual Rhythm

TL;DR

This paper introduces a novel video-based ALPR system that efficiently extracts a single representative frame per vehicle and recognizes license plates using OCR, reducing computational load while maintaining accuracy.

Contribution

The proposed system uniquely captures only one frame per vehicle for license plate recognition, improving efficiency over traditional multi-frame methods.

Findings

Viability demonstrated through early experiments

Reduces processing time by using single-frame extraction

Maintains recognition accuracy with simplified approach

Abstract

Automatic License Plate Recognition (ALPR) involves extracting vehicle license plate information from image or a video capture. These systems have gained popularity due to the wide availability of low-cost surveillance cameras and advances in Deep Learning. Typically, video-based ALPR systems rely on multiple frames to detect the vehicle and recognize the license plates. Therefore, we propose a system capable of extracting exactly one frame per vehicle and recognizing its license plate characters from this singular image using an Optical Character Recognition (OCR) model. Early experiments show that this methodology is viable.