A Bus Crowdedness Sensing System Using Deep-Learning Based Object Detection

Abstract

Crowdedness of buses is playing an increasingly important role in the disease control of COVID-19. The lack of a practical approach to sensing the crowdedness of buses is a major problem. This paper proposes a bus crowdedness sensing system which exploits deep learningbased object detection to count the numbers of passengers getting on and off a bus and thus estimate the crowdedness of buses in real time. In our prototype system, we combine YOLOv5s object detection model with Kalman Filter object tracking algorithm to implement a sensing algorithm running on a Jetson nano-based vehicular device mounted on a bus. By using the driving recorder video data taken from real bus, we experimentally evaluate the performance of the proposed sensing system to verify that our proposed system system improves counting accuracy and achieves real-time processing at the Jetson Nano platform.

A Driver Fatigue Detection Algorithm Based on Dynamic Tracking of Small Facial Targets Using YOLOv7

“…Chen et al [19] introduced a dual-task fusion network that performs monitor and segment tasks on pedestrians, outperforming many state-of-theart detectors. Huang et al [20] combined the YOLOv5s target detection model with a Kalman filter target tracking algorithm, via a mounted Jetson Nano in-vehicle device to count the number of passengers on the bus. In a different application, Liu et al [21] proposed a CEAM-YOLOv7 algorithm based on channel expansion and attention mechanisms for driver distracted behavior detection, achieving a significant improvement of 20.26% in mAP and 156 FPS.…”