This content is not included in your SAE MOBILUS subscription, or you are not logged in.
Road Crossing Assistance Method Using Object Detection Based on Deep Learning
ISSN: 0148-7191, e-ISSN: 2688-3627
Published March 29, 2022 by SAE International in United States
Annotation ability available
This paper describes a method for assisting pedestrians to cross a road. As motorization develops, pedestrian protection techniques are becoming more and more important. Advanced driving assistance systems (ADAS) are improving rapidly to provide even greater safety. However, since the accident risk of pedestrians remains high, the development of an advanced walking assistance system for pedestrian protection may be an effective means of reducing pedestrian accidents. Crossing a road is one of the highest risk events, and is a complex phenomenon that consists of many dynamically changing elements such as vehicles, traffic signals, bicycles, and the like. A road crossing assistance system requires three items: real-time situational recognition, a robust decision-making function, and reliable information transmission. Edge devices equipped with autonomous systems are one means of achieving these requirements. Situational recognition when crossing a road must identify the pedestrian traffic signals and the crosswalk. Various research has been published regarding the recognition of vehicle traffic signals and crosswalks using in-vehicle cameras. However, since crosswalks (and pedestrians) have conventionally been treated as risks or obstacles for vehicles rather than as guides for crossing, these recognition methods cannot be diverted directly for pedestrian assistance. This paper proposes a novel methodology for walking assistance that includes an image recognition system based on a combination of convolute neural network (CNN) and computational visualization technologies (CV). The proposed methodology also includes a robust judgment algorithm for crossing roads. The proposed method is implemented on edge devices, and its efficacy has been confirmed in field tests. The developed system is considered to be effective and efficient for providing walking assistance.
CitationKawamura, H., Shintani, K., Mima, H., and Taniguchi, M., "Road Crossing Assistance Method Using Object Detection Based on Deep Learning," SAE Technical Paper 2022-01-0149, 2022, https://doi.org/10.4271/2022-01-0149.
- World Health Organization Global Status Report on Road Safety 2018 2018 978241565684
- Cades , D.M. , Crump , C. , Lester , B.D. , and Young , D. Driver Distraction and Advanced Vehicle Assistive Systems (ADAS): Investigating Effects on Driver Behavior Advances in Human Aspects of Transportation 2016 1015 1022 10.1007/978-3-319-41682-3_83
- Wen , L. , Bastien , C. , Blundell , M. , Neal-Sturgess , C. et al. Stability and Sensitivity of THUMS Pedestrian Model and its Trauma Response to a Real Life Accident 10th European LS-DYNA Conference 2015 2015
- Japan National Police Agency 2021 https://www.npa.go.jp/publications/statitics/koutuu/toukeihyo_e.html
- National Highway Traffic Safety Administration 2021 https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813079
- Wonghabut , P. , Kumphong , J. , Ung-arunyawee , R. , Leelapatra , W. et al. Traffic Light Color Identification for Automatic Traffic Light Violation Detection System 2018 International Conference on Engineering, Applied Sciences, and Technology (ICEAST) 1 4 2018 10.1109/ICEAST.2018.8434400
- Tumen , V. and Ergen , B. Intersections and Crosswalk Detection Using Deep Learning and Image Processing Techniques Physica A: Statistical Mechanics and its Applications 543 2019 10.1016/j.physsa.2019.123510
- Zhai , Y. , Cui , G. , Qin , G. , and Kong , L. Crosswalk Detection based on MSER and ERANSAC IEEE Int. Conf. on Intelligent Transportation Systems 2015 10.1109/ITSC.2015.448
- Bochkovskiy , A. , Wang , C.Y. , and Liao , H.Y.M. 2020
- Shorten , C. and Khoshgoftaar , T.M. A Survey on Image Data Augmentation for Deep Learning Journal of Big Data 6 2019 10.1186/s40537-019-0197-0