DOI: 10.18178/wcse.2020.02.022
A Study On Deep Learning Based Real Time Road information Monitoring Device for Emergency Vehicle Guidance
Abstract— This study is to secure the golden time for emergency vehicles, especially to shorten time in narrow roads. These days, in Korea the black box install rate is over 88.9% and it shows that Koreans use black box much more than other countries. Through this study we can analyze the road situation by collecting the data through the black box and transmit it to the server. And not just guiding the emergency vehicle to the shortcut but to the optimal route when they pass through alleys. Finally we could suggest the solution for the emergency vehicle to be in golden time. For this study we used Raspberry-Pi and Yolo v3 algorithm. There were difference between analyzing the image and the video and concluded that the video take much more time due to the transmission delay and data size.
Index Terms— Deep Learning, Emergency Vehicle Guidance, Real Time Road Information, Yolo v3
Seona Park, Junghan Ha, Muwook Pyeon, Wonwoo Jung
Konkuk University, REPUBLIC OF KOREA
Cite: Seona Park, Junghan Ha, Muwook Pyeon, Wonwoo Jung, "A Study On Deep Learning Based Real Time Road information Monitoring Device for Emergency Vehicle Guidance," Proceedings of 2020 the 10th International Workshop on Computer Science and Engineering (WCSE 2020), pp. 130-135, Yangon (Rangoon), Myanmar (Burma), February 26-28, 2020.