SECURITY FRAMEWORK FOR CYBER-PHYSICAL SMART PARKING SYSTEMS WITH AUTOMATED LICENSE PLATE RECOGNITION

Abstract

Smart parking systems with automated license plate recognition (ALPR) are getting more popular in cities, but they have serious cybersecurity problems. This study analyzes security threats in smart parking infrastructure and offers ways to reduce them via vulnerability assessments and improved security designs. We analyzed various attacks such as protocol exploits and data interception risks in cyber-physical parking systems. Our research studies RTSP camera communication vulnerabilities, REST API security problems, and cloud service integration risks in license number recognition systems using computer vision technologies. Our approach includes vulnerability testing, threat modeling with STRIDE framework, penetration testing, and security analysis. We studied problems of RTSP camera protocol, HTTP/HTTPS communications, Laravel REST API setup, and Google Cloud Vision API integration. Results show that smart parking systems may have data interception risks, unauthorized access, API security problems and system integrity threats, which need multi-layered security approaches. We designed a cyber-physical parking system prototype with improved security measures in all components. The prototype has good license plate recognition accuracy while applying security methods without major performance reduction. Important security factors include secure communication protocols, encrypted transmission of data, authentication frameworks, input validation, rate limiting, and logging systems. This research helps to understand cybersecurity facets in IoT-based parking systems and offers methods for secure automated vehicle recognition setup in smart cities. This work is truly relevant for Ukrainian smart city projects, showing methodology applicable by IT companies for critical infrastructure protection.