Recognition of driving postures by contourlet transform and random forests

An efficient feature extraction approach for driving postures from a video camera, consisting of Homomorphic filtering, skin-like regions segmentation and contourlet transform (CT), was proposed. Features were extracted from a driving posture dataset created at Southeast University (SEU), then holdout and cross-validation experiments on driving posture classification were conducted using random forests (RF) classifier. Compared with a number of commonly used classification methods including linear perceptron classifier, k-nearest-neighbour classifier and multilayer perceptron (MLP) classifier, the experiment results showed that the RF classifier offers the best classification performance among the four classifiers. Among the four predefined classes, that is, grasping the steering wheel, operating the shift gear, eating and talking on a cellular phone - the class of eating is the most difficult to classify. With RF classifier, the classification accuracies of eating are over 88% in holdout and cross-validation experiments, demonstrating the effectiveness of the proposed feature extraction method and the importance of RF classifier in automatically understanding and characterising driver’s behaviours towards human-centric driver assistance systems.

• Record URL:
  • https://doi.org/10.1049/iet-its.2011.0116
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/1751956X
• Authors:
  • Zhao, C H
  • Zhang, B L
  • He, Jianhong
  • Lian, J
• Publication Date: 2012-6

Language

• English

Media Info

• Media Type: Digital/other
• Features: Figures; References; Tables;
• Pagination: pp 161-168
• Serial:
  • IET Intelligent Transport Systems
  • Volume: 6
  • Issue Number: 2
  • Publisher: Institution of Engineering and Technology (IET)
  • ISSN: 1751-956X
  • EISSN: 1751-9578
  • Serial URL: https://ietresearch.onlinelibrary.wiley.com/journal/17519578
• Publication flags:

  Open Access (libre)

Subject/Index Terms

• TRT Terms: Alertness; Automobile drivers; Behavior; Data collection; Driver support systems; Image analysis; Pattern recognition systems; Posture; Video cameras
• Subject Areas: Data and Information Technology; Highways; Passenger Transportation; Vehicles and Equipment; I91: Vehicle Design and Safety;

Filing Info

• Accession Number: 01375692
• Record Type: Publication
• Files: TRIS
• Created Date: Jul 18 2012 4:11PM