Congratulations to our MSAAI graduate, Li-Sha Zhou, on having her master’s capstone project accepted at the International Conference on Machine Intelligence and Nature-Inspired Computing 2025 (https://www.ic-mind.org/)!
- Li-Sha Zhou, Richard Tai-Chiu Hsung, Harris Sik-Ho Tsang, Tony Yu-Lin Zhu, Wai-Lun Lo, “A Robust Drone based Parking Space Detection System Based on Deep Learning Algorithms,” 2025 The International Conference on Machine Intelligence and Nature-inspireD Computing (MIND), Xiamen, China, 2025, (accepted).
Paper Abstract
As urbanization and vehicle numbers rise, traditional parking management struggles to meet the demands of modern cities, especially in China. Existing parking detection technologies, dependent on sensors like geomagnetic and infrared, face high costs and maintenance challenges. However, many urban areas utilize surveillance cameras, providing a foundation for advanced computer vision systems. This paper presents a novel solution using high-resolution drone video and publicly available datasets to detect parking space availability in real time with support for bird and oblique views. Employing the YOLOv8 model, optimized for various conditions, the system demonstrates impressive accuracy, achieving precision, recall, and mAP scores of 0.97, 0.90, and 0.94, respectively. This research underscores the potential of deep learning in addressing urban parking challenges, offering a scalable and cost-effective solution that can be integrated into existing infrastructure. Moving forward, the system’s integration with autonomous vehicles and smart city initiatives could pave the way for more intelligent, efficient urban environments.
Some photos of the paper
Detection results of some pictures and videos.
部分圖片和影片的檢測結果。
