Character Time-series Matching For Robust License Plate Recognition

TL;DR

This paper introduces a multi-frame license plate recognition method that improves accuracy under challenging conditions by tracking and matching characters over time, achieving high performance on datasets and real-world deployment.

Contribution

It proposes Character Time-series Matching and Adaptive License Plate Rotation algorithms to enhance ALPR accuracy in real-world scenarios with variable conditions.

Findings

96.7% accuracy on UFPR-ALPR dataset

0.881 license plate detection mAP

0.979 character recognition mAP

Abstract

Automatic License Plate Recognition (ALPR) is becoming a popular study area and is applied in many fields such as transportation or smart city. However, there are still several limitations when applying many current methods to practical problems due to the variation in real-world situations such as light changes, unclear License Plate (LP) characters, and image quality. Almost recent ALPR algorithms process on a single frame, which reduces accuracy in case of worse image quality. This paper presents methods to improve license plate recognition accuracy by tracking the license plate in multiple frames. First, the Adaptive License Plate Rotation algorithm is applied to correctly align the detected license plate. Second, we propose a method called Character Time-series Matching to recognize license plate characters from many consequence frames. The proposed method archives high performance in the UFPR-ALPR dataset which is \boldmath$96.7\%$ accuracy in real-time on RTX A5000 GPU card. We also deploy the algorithm for the Vietnamese ALPR system. The accuracy for license plate detection and character recognition are 0.881 and 0.979 $mAP^{test}$@.5 respectively. The source code is available at https://github.com/chequanghuy/Character-Time-series-Matching.git