The rapid growth of urbanization has intensified the demand for efficient parking management solutions, particularly in large indoor commercial facilities, where conventional sensing infrastructures struggle with installation complexities and maintenance overheads. This study presents an enhanced SPS that leverages SBMA for parking slot occupancy detection using live video streams. Unlike sensor-based or computationally intensive deep learning approaches, the SBMA provides a lightweight and robust statistical comparison framework suitable for constrained hardware environments. The system supports both SCSL and MCML deployments, enabling scalable monitoring across parking zones of different sizes. Integrated graphical and mobile interfaces offer real-time visualization and user guidance. Experiments conducted on real-world data from two shopping malls demonstrated that the proposed approach achieved reliable occupancy classification under illumination variations and moderate occlusions while providing competitive processing times. The results indicate that the SBMA-driven SPS with a PSNR value and processing time is a practical and cost-effective tool for smart city parking infrastructure.