DLSU Senior High School Research Congress Conference Proceedings
Document Type
Paper Presentation
Research Advisor (Last Name, First Name, Middle Initial)
De Guia, Leovine L.
Abstract/Executive Summary
Complex challenges in parking have been prevalent in metropolitan areas worldwide due to factors such as inefficient use of space and rapid urbanization, which in turn exacerbate traffic conditions and potentially lead to environmental issues. While traditional efforts, such as increasing parking infrastructure, have been made, this study focuses on developing a smart parking reservation system that integrates camera-based software for enhanced parking management. With the use of CCTV camera technology, the detection system was tested under various lighting scenarios to evaluate its efficiency in terms of accuracy and consistency. Developed under a developmental research design, a parking reservation website was created, coinciding with the evaluation of the system’s ability to detect vacant and non-vacant spaces. The results showed that both precision and recall began at 91.56% and 90.06%, respectively, during the early morning hours and steadily increased throughout the day, surpassing 96.9% precision and 95.6% recall by mid-afternoon. This enhancement in detection accuracy is consistent with previous studies, which indicate that YOLO-based detection systems achieve their best performance in well-lit environments. Although the system successfully met the target threshold of ≥ 90%, tests conducted during nighttime revealed that factors such as low light, glare, and the presence of other objects continue to present challenges. Overall, the research highlights that an affordable, camera-based smart parking solution can substantially improve parking management, maximize space utilization, and elevate the user experience.
Keywords
CCTV; parking reservation system; IoT; camera detection; YOLO