Vehicle Occupant Head Distance Extraction Method using Stereo-Vision

In recent years, the development of airbag system has improved the safety of the occupants. However, the car occupant can be often killed by improper airbag operation. Robust detection of vehicle occupant posture is necessary for a safer intelligent airbag deployment system. This paper presents a stereobased method of estimating the posture of the occupant. Utilizing the disparity images and the accumulated difference image, the SAD is executed in stereo images within the candidate region. Our experiments show that the property of stereo geometry is useful to detect and classify head position with the disparity between stereo images.

[1]  Alexandros Eleftheriadis,et al.  Automatic face location detection for model-assisted rate control in H.261-compatible coding of video , 1995, Signal Process. Image Commun..

[2]  Narendra Ahuja,et al.  Detecting Faces in Images: A Survey , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  S.J. Krotosky,et al.  Occupant posture analysis using reflectance and stereo image for "smart" airbag deployment , 2004, IEEE Intelligent Vehicles Symposium, 2004.

[4]  Mohan M. Trivedi,et al.  Occupant posture analysis with stereo and thermal infrared video: algorithms and experimental evaluation , 2004, IEEE Transactions on Vehicular Technology.

[5]  Kurt Konolige,et al.  Small Vision Systems: Hardware and Implementation , 1998 .

[6]  S.Y. Cheng,et al.  Human posture estimation using voxel data for "smart" airbag systems: issues and framework , 2004, IEEE Intelligent Vehicles Symposium, 2004.

[7]  Gwi-Tae Park,et al.  Vision Based Automatic Occupant Classification and Pose Recognition for Smart Airbag Deployment , 2005, EUROCAST.

[8]  Takeo Kanade,et al.  A stereo machine for video-rate dense depth mapping and its new applications , 1996, Proceedings CVPR IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[9]  H. Bischof,et al.  Robust occupancy detection from stereo images , 2004, Proceedings. The 7th International IEEE Conference on Intelligent Transportation Systems (IEEE Cat. No.04TH8749).