PERSONAL MOBILITY DETECTION USING YOLO DEEP LEARNING ALGORITHM

Abstract

Recently, the utilization rate of Personal Mobility (PM) and its users have been rapidly increasing as a short-distance transportation option. As the consumption patterns shifted towards the sharing economy, various shared mobility platforms have been developed, leading to the emergence of PM in the form of shared electric scooters. Consequently, there has been a simultaneous increase in companies providing shared PM services. However, due to the diversity of shared PM offered by different service providers and variations in the number of providers across regions, the comprehensive management of PMs has become more challenging. Therefore, this paper aims to evaluate the feasibility of utilizing the YOLOv3 algorithm to detect shared PM objects from images collected by drones. The detection accuracy was evaluated to verify the potential for integrated management of PMs. PM objects within the experimental area were collected by drones, and labeling was performed to train a deep learning model for PM detection. The experimental results demonstrated a detection accuracy of 87.38% and an AP(average precision) value of 0.73, indicating high viability of utilizing the YOLOv3 algorithm on drone images to detect shared PMs. © 2023 ACRS. All Rights Reserved.