IoT-Based Automatic Parking Monitoring System Using Computer Vision Technology

Abstract

State Polytechnic of Madiun faces challenges in managing parking areas, particularly during large events that lead to increased vehicle volume. The absence of information regarding parking space availability causes congestion and inconvenience for users. To address this issue, an Internet of Things (IoT)-based smart parking system integrated with computer vision technology using the YOLOv8n algorithm was developed.The system consists of an IP Camera and Raspberry Pi 4B to detect vehicles in real time, supported by a website interface connected to Firebase Realtime Database for displaying parking slot status in real time. This study employed the Research and Development (R&D) method with a prototyping approach, involving stages of requirement gathering, prototype development and evaluation, system development, system testing and evaluation, and system implementation. Testing was conducted using the black-box testing method to evaluate detection accuracy and overall system performance. The results indicate that the system successfully detected vehicles with satisfactory accuracy, although limitations were identified in Raspberry Pi performance, including delays and reduced frames per second (FPS) when detecting multiple objects simultaneously. Occupied parking slots are displayed in red, while available slots are shown in green. Detection results are stored in Firebase and presented in real time through a website featuring parking information, an admin dashboard, parking and staff data management, and user profile functionality. The implementation of this system provides an effective solution for monitoring parking conditions and supports more structured parking area management at Campus 2 of the State Polytechnic of Madiun.