Fine Grained abnormal driving

Abstract

period abnormal driving behaviors observation might be an anchor to improving driving safety. Existing works on
driving behaviors observation using smart phones only supply a coarse-grained result, i.e. characteristic abnormal driving
behaviors from normal ones. to boost drivers’ awareness of their driving habits therefore on stop potential car accidents, we would
like to think about a fine-grained observation approach, that not only detects abnormal driving behaviors but in addition identifies
specific varieties of abnormal driving behaviors, i.e. Weaving, Swerving, side slippery , fast reversion, Turning with a large radius
and sudden braking. Through empirical studies of the 6-month driving traces collected from real driving environments, we tend to
discover that everyone among the six forms of driving behaviors have their distinctive patterns on acceleration and orientation.
Recognizing this observation, we tend to further propose a fine-grained abnormal Driving behavior Detection and identification
system to perform real-time high-accurate abnormal driving behaviors observation using smart phone sensors. We tend to extract
effective choices to capture the patterns of abnormal driving behaviors. After that, a pair of machine learning ways, rash driving,
or officially driving under the Influence (DUI) of alcohol, can be a serious reason for traffic accidents throughout the globe. In
this, we tend to tend to propose a extraordinarily efficient system geared toward early detection and alert of dangerous vehicle
maneuvers typically related to rash driving. The whole solution wants only a mobile placed in vehicle and with accelerometer
device. A program installed on the mobile automatically computes accelerations supported detector readings, and compares them
with typical rash driving patterns extracted from real driving tests. Once any proof of rash driving is present, the mobile will
automatically alert the driver or sends a message to predefined number in application for help well before accident actually
happens.