Smart Parking System with Space Detection Using Deep Learning in a Mobile Application

View at publisher Summarise with VIRAT
Authors: Anis Amilah Shari, Mohd Rahmat Mohd Noordin, Asma Shazwani Shari, Asiah Mat, Farah Muna Mohamad Ghazali, Siti Anisah Mohd Hatta, Norlida Binti Razali, Naufal Anas Shafie
Cited by: 0
References: 8

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Parking congestion in university areas remains a persistent challenge, leading to extended search times, fuel waste, and frustration among students, staff, and visitors. Traditional campus parking systems lack real-time availability tracking and automated space allocation, resulting in these inefficiencies. This paper presents a smart parking system specifically designed for university campuses, integrating deep learning-based YOLOv8n for parking space detection and Dijkstra's algorithm for optimal route allocation. The system features a Flutter-based mobile application connected to a Flask backend, enabling users to view real-time parking availability, receive automatic assignments to the nearest available space, and navigate directly to their assigned spot. A custom dataset combining the PKLot dataset with campus- specific images was curated, preprocessed, annotated, and used for training and validation. The optimized YOLOv8n model achieved a mean Average Precision ($\text{mAP} {@} 0.5$) of 0.994, precision of 0.998, and recall of 0.998. Experimental results demonstrate that the system significantly reduces parking search time, optimizes space utilization, and enhances user convenience. Future enhancements may include multi-zone coverage expansion, license plate recognition for security, and predictive analytics for peak-hour availability estimation.

Fields of Study & Subjects

Smart Parking Systems Research Advanced Neural Network Applications Vehicle License Plate Recognition

Publication details

Journal / source
International Conference on Cognitive Systems and Computer Interaction (ICoSCI)
Pages
1–4
Publication date
15 January 2026
Language
EN