Enhanced Detection of Illegally Parked Vehicles Using YOLO and Good Feature to Track Methods

The study addresses illegal parking, a problem that disrupts traffic and poses safety risks under Indonesia’s legal system. It proposes using the YOLO technique for detection and Optical Flow for vehicle speed assessment to identify illegally parked vehicles. The objective is to improve detection precision and efficiency in areas prone to illegal parking by leveraging deep learning, particularly Convolutional Neural Networks (CNN). The study highlights the potential of noninvasive computer vision technology to enhance public awareness and law adherence. The findings aim to support the development of a monitoring system to help law enforcement in managing illegal parking in public spaces.