Conference Paper

A Novel Driving Pattern Recognition and Status Monitoring System.

DOI: 10.1007/11949534_50 Conference: Advances in Image and Video Technology, First Pacific Rim Symposium, PSIVT 2006, Hsinchu, Taiwan, December 10-13, 2006, Proceedings
Source: DBLP

ABSTRACT This paper describes a novel driving pattern recognition and status monitoring system based on the orientation information.
Two fixed cameras are used to capture the driver’s image and the front-road image. The driver’s sight line and the driving
lane path are found from these 2 captured images and are mapped into a global coordinate. Two correlation coefficients among
the driver’s sight line, the driving lane path and the car heading direction are calculated in the global coordinate to monitor
the driving status such as a safe driving status, a risky driving status and a dangerous driving status. The correlation coefficients
between the lane path and car heading direction in a fixed period are analyzed and recognized as one of 4 driving patterns
by HMM. Four driving patterns including the driving in a straight lane, the driving in a curve lane, the driving of changing
lanes, and the driving of making a turn are able to be recognized so far.

0 Followers
 · 
86 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: Road traffic accidents (RTAs) pose a public health and development challenge and greatly affect the human capital development of every nation. Main causes for a significant percentage of RTAs are recklessness, fatigue or stress, inexperienced driving and driving under the influence of alcohol. Reputed automobile companies and authorities have established standards and/or proactive/reactive solutions over the years to overcome this problem but the rates continue to grow each year. Software solutions that focus on preventing RTAs are difficult to be found in economy class automobiles that are used by the majority of consumers. The limited number of successful applications are either bound to a certain brand/model and/or focus on a specific task. SmartV is smart phone based vigilance monitoring system that focuses on the majority of factors contributing to RTAs. These include DUI (driving under the influence of alcohol), health issues and vehicle defects. The mobile application communicates with a heart rate monitor and a Bluetooth OBDII Adapter. These will continuously monitor the heart rate of the driver and the status of the vehicle respectively. During the course of the drive, the system monitors the behavior of the vehicle for existence of dangerous driving patterns that are defined in Visual Detection of DWI Motorists[8] a study conducted by U.S.A. N.H.T.S.A. The final product is a versatile, non-intrusive, flexible and most importantly an affordable mobile application that captures the vehicle's behavior, the health level of the driver and the status of the vehicle. Real environment operation of SmartV has yielded significant results proving itself to be a successful and an affordable Vigilance Monitoring System for the future.
    2013 IEEE 8th International Conference on Industrial and Information Systems (ICIIS); 12/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: In view of the shortcoming of driver state detection system which may take turn behavior as driver distraction, a new method to recognize turn behavior was proposed based on video images analysis. The driver hands position in different driving behavior were analyzed and we found that the position of driver’s hands changed more violent when in turning than in other driving behavior. So we may use standard deviation of driver hands position to recognize driver turn behavior. In order to improve the hands locating speed the Particle Filtering was used to track the driver hands. And experiments resulted that the recognition algorithm can identify the driver's turn behavior.
    01/2012; 433-440:6230-6234. DOI:10.4028/www.scientific.net/AMR.433-440.6230
  • [Show abstract] [Hide abstract]
    ABSTRACT: In recent year many researcher and industries are working on VANET and trying to implement the concepts in real world. Many VANET systems are proposed and tested on simulation but very few inventors implemented it. Vehicular ad hoc networks (VANETs) are being developed to provide on-demand wireless communication infrastructure among vehicles and authorities. Such an infrastructure is expected to deliver multiple road safety and driving assistance applications. Vehicles will be equipped with sensors and communication devices that will allow them to cooperate with each other. Vehicles can exchange different type of information as per requirements on demand for specified application. With the purpose of supporting and improving data collection and distribution, in this paper a smart phone-based platform is designed that exploits low-cost dedicated hardware to interact with sensors on board and in the vehicle surroundings.
    2014 International Conference on Communication Systems and Network Technologies (CSNT); 04/2014