SafeDrive: Online Driving Anomaly Detection From Large-Scale Vehicle Data

Zhang, Mingming; Chen, Chao; Wo, Tianyu; Xie, Tao; Bhuiyan, Zakirul Alam; Lin, Xuelian

doi:10.1109/tii.2017.2674661

Cited by 108 publications

(47 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Zhang et al [23] have built a system called SafeDrive, to detect abnormal driving behaviors from large-scale vehicle data State Graph (SG). The parameters used are hard acceleration and hard braking.…”

Section: Pattern Monitoring Using Obdmentioning

confidence: 99%

A survey on driving behavior analysis in usage based insurance using big data

Arumugam

Bhargavi

2019

J Big Data

View full text Add to dashboard Cite

An accident is defined as an unfortunate incident that happens unexpectedly and unintentionally, typically resulting in damage or injury. Considering all the consequences that could eventuate after an accident, there are reasons to believe that a normal person does not drive with an ex-ante intention to cause an accident. Holding a valid driving license is prerequisite to drive in any part of the world and during the licensing

show abstract

Section: Pattern Monitoring Using Obdmentioning

confidence: 99%

A survey on driving behavior analysis in usage based insurance using big data

Arumugam

Bhargavi

2019

J Big Data

View full text Add to dashboard Cite

show abstract

“…Peng, Li, Li, and Zhu proposed a platform integrated with distributed storages and parallel computing technologies for traffic data management and analysis [15]. Based on the well-managed large amount of data by cloud computing technologies, further traffic or driving analysis can be achieved [16]. Finally, Dong et al presented an intelligent framework based on IoT, 4G, big data, cloud computing, and other novel ICT technologies [17].…”

Section: B Intelligent Transport Systems and Cloud Computingmentioning

confidence: 99%

Cloud-Assisted Safe Driving Framework

Kao¹

2017

IJMO

View full text Add to dashboard Cite

Abstract-As the technologies for vehicular safety grow and mature, several advanced driving assisting systems are deployed to mitigate negative impacts on driving safety and protect drivers and passengers efficiently. However, the rate of fatalities and serious injuries in traffic accidents is still high. Based on the statistics from the Ministry of the Interior and National Highway Police Bureau in Taiwan, most critical accidents are caused by the circumstance that drivers do not aware the potential danger to their surroundings. Currently, several driving assisting systems deployed in vehicles can sense surrounding information for potential accidents prevention in time. In order to prevent potential accidents in advance for safer driving, a cloud-assisted safe driving framework is proposed in this paper. The framework leverages sensing devices in vehicles to acquire particular information of the front vehicles. The information will be acquired, recognized, and analyzed based on historical data preserved in the cloud. In addition, the information and comments on social network sites based on the recognized information will be collected and analyzed. Based on historical data and social information, a reputation for the recognized vehicle will be calculated in the cloud. Thus, corresponding suggestion based on the reputation is given to the drivers to perform defensive driving if necessary. Through the cloud-assisted safe driving framework, the driving safety can be achieved in advance efficiently.

show abstract

“…We would like to emphasise that even though we are demonstrating our approach on three different applications specifically in the medical domain, the varied nature of these problems demonstrates how the proposed model can be directly applied to any anomaly detection problem in different domains where modelling long term relationships is necessary. Possible application areas include, detecting anomalies in daily human activities and sports activities [16], anomaly detection in vehicle driving [17], and detecting anomalies in stock exchange [18] and in credit card transactions [19].…”

Section: Introductionmentioning

confidence: 99%