Real-Time Illegal Parking Detection System Based on Deep Learning
This addresses illegal parking detection for urban management, but it is incremental as it builds on existing SSD methods with minor optimizations.
The paper tackled the problem of detecting illegally parked vehicles by proposing a deep learning-based system that uses an optimized SSD algorithm for detection and tracking for analysis, achieving 99% accuracy and real-time performance at 25 FPS.
The increasing illegal parking has become more and more serious. Nowadays the methods of detecting illegally parked vehicles are based on background segmentation. However, this method is weakly robust and sensitive to environment. Benefitting from deep learning, this paper proposes a novel illegal vehicle parking detection system. Illegal vehicles captured by camera are firstly located and classified by the famous Single Shot MultiBox Detector (SSD) algorithm. To improve the performance, we propose to optimize SSD by adjusting the aspect ratio of default box to accommodate with our dataset better. After that, a tracking and analysis of movement is adopted to judge the illegal vehicles in the region of interest (ROI). Experiments show that the system can achieve a 99% accuracy and real-time (25FPS) detection with strong robustness in complex environments.