Abstract
In recent years, a drive recorder becomes common and is installed in a car to record sensor data, such as images, acceleration, and speed, about driving. The recorded data is useful to confirm and analyze a dangerous driving scene of a traffic accident and an incident. However, analyzing such data takes long time because it is done by a person who checks data one by one. Therefore, a method of automatic classification of drive recorder data is explored in this study. First, we labeled three types of incidents on the recorded data. Then, after extracting features from the acceleration and velocity, machine learning techniques are applied for the classification. Our preliminary evaluation showed that the classification result achieved about 0.55 of f-measure value.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Safety measures for business vehicles: vehicle comprehensive safety information (in Japanese). http://www.mlit.go.jp/jidosha/anzen/03driverec/dorareco.html
Kubo, N., Midori, M.: Efficient automatic classification method of drive recorder. Trans. Jpn. Soc. Mech. Eng. Ser. (C) 77(778), 338–353 (2011). (in Japanese)
Takenaka, K., Bando, T., Nagasaka, S., Taniguchi, T.: Drive video summarization based on double articulation structure of driving behavior. In: MM 2012 Proceedings of the 20th ACM International Conference on Multimedia, pp. 1169–1172 (2012)
News September 26, 2016: Succeeded in automatic detection of dangerous driving using artificial intelligence (AI) – NTT Com Corporate Information, NTT Communications Company Information (in Japanese). http://www.ntt.com/about-us/press-releases/news/article/2016/20160926_2.html
NTT Com: Successful automatic detection of dangerous driving using AI (in Japanese). http://www.atmarkit.co.jp/ait/articles/1609/27/news117.html
NTT Com and NCS: 96% probability to detect a temporary stop violation from video of drive recorder (in Japanese). https://it.impressbm.co.jp/articles/-/15087
Traffic Accident Statistics Glossary (in Japanese). https://www.itarda.or.jp/service/term.php
Japanee Industrial Standards JIS D 5601 – 1992 Automobiles speedometer (in Japanese)
Weka 3 - Data Mining with Open Source Machine Learning Software in Java. https://www.cs.waikato.ac.nz/ml/weka/
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering
About this paper
Cite this paper
Miyata, M., Matsuo, K., Omura, R. (2018). Automatic Classification of Traffic Accident Using Velocity and Acceleration Data of Drive Recorder. In: Murao, K., Ohmura, R., Inoue, S., Gotoh, Y. (eds) Mobile Computing, Applications, and Services. MobiCASE 2018. Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering, vol 240. Springer, Cham. https://doi.org/10.1007/978-3-319-90740-6_17
Download citation
DOI: https://doi.org/10.1007/978-3-319-90740-6_17
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-90739-0
Online ISBN: 978-3-319-90740-6
eBook Packages: Computer ScienceComputer Science (R0)