Visual Detection Technology of the Driver's State of Fatigue Based on Multi-Information Fusion

Abstract:

This paper attempts to combine the image features of shape, movement and color, taking the oval as the shape model of the target to immediately catch the edge of the target, using the feature light stream technology to obtain the movement information of the target. The color information can be obtained using the color histogram features of the target and mean shift algorithm. In the tracking of the color, SDA color spatial selection algorithm is introduced to improve the target and background differentiation. Then, the above information would be integrated through particle filtering technology. The self-adapting integration technology is used, and each image features weight would be decided according to its validity, thus, a quick and robust facial tracking algorithm is gained.

• Periodical: Applied Mechanics and Materials (Volumes 380-384)
• Pages: 3925-3929
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.380-384.3925
• Online since: August 2013
• Authors: Li Ming Zhao, Yi Min Zou
• Keywords: Multi-Feature Information, Oval Shape-Changing Model, Particle Filtering Wave, Self-Adapting Integration Tracking Algorithm

References

[1] Bascle B, Deriche R. Region Tracking Through Image Sequences[C] Proceedings of Fifth IEEE International Conference on Computer Vision (ICCV), Cambridge, Ma, USA: IEEE Computer Society Press. (1995).

DOI: 10.1109/iccv.1995.466925

[2] Scaggiante A, Frezza R, Zampato M. Identifying and Tracking Ellipses: a technique based on elliptical active contours[C]/Proceedings of 10th International Conference on Image Analysis and Processing. Venice, Italy: IEEE Computer Society Press. (1999).

DOI: 10.1109/iciap.1999.797659

[3] E. Maggio, F. Smeraldi, A. Cavallaro, Adaptive multi-feature tracking in a particle filtering framework', IEEE Trans. on Circuits and Systems for Video Technology, Vol. 17, Issue 10, Oct. 2007, pp.1348-1359.

DOI: 10.1109/tcsvt.2007.903781