Head detection inside vehicles with a modified SVM for safer airbags

Support vector machines have shown great performances in face recognition applications in nonconstraint images. We propose to modify the threshold of the SVM decision function in order to add some spatial information for head recognition. We propose this method as a passenger occupancy detection system. The information about the presence of a person in the passenger seat and the position of his head is used to control the deployment of the passenger airbag. We use still monocular images taken from the cockpit to make the learning step of the SVM method. During recognition the threshold of the decision function will be modified by a "presence" probability distribution function. We will show in this paper the results of our simulations for a standard and for the modified SVM.

[1]  D. Houzet,et al.  Implementation of the SVM neural network generalization function for image processing , 2000, Proceedings Fifth IEEE International Workshop on Computer Architectures for Machine Perception.

[2]  Massimiliano Pontil,et al.  Properties of Support Vector Machines , 1998, Neural Computation.

[3]  Serge Boverie,et al.  INSIDE VEHICLE PERCEPTION FOR SAFETY APPLICATIONS , 2000 .

[4]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[5]  M.M. Van Hulle,et al.  View-based 3D object recognition with support vector machines , 1999, Neural Networks for Signal Processing IX: Proceedings of the 1999 IEEE Signal Processing Society Workshop (Cat. No.98TH8468).

[6]  M. Devy,et al.  Detection and classification of passenger seat occupancy using stereovision , 2000, Proceedings of the IEEE Intelligent Vehicles Symposium 2000 (Cat. No.00TH8511).