Automated Enforcement of Vehicle Tail Number Restrictions via SVM and Bayesian Optimization

Abstract

With urban congestion intensifying, vehicular tail number restriction policies have emerged as an efficacious strategy to mitigate traffic burden. Traditional enforcement, hinging on manual verification, suffers from inefficiencies and vulnerability to human error. Addressing these drawbacks, the present study introduces a Support Vector Machine (SVM)-based license plate recognition technique, augmented by Bayesian optimization, to automate adherence to tail number restrictions. Utilizing the Bayesian optimization algorithm for the fine-tuning of SVM parameters enables the proposed system to identify license plate numbers with heightened precision and swiftness, thus streamlining the enforcement process. Comparative experimental analysis substantiates that our proposed model outperforms existing methods, demonstrating superior recognition efficiency and accuracy. This advancement heralds a significant leap towards the integration of intelligent systems in urban traffic management.