Wavelet-based multispectral face recognition

Abstract

This paper proposes a novel wavelet-based face recognition method using thermal infrared (IR) and visible-light face images. The method applies the combination of Gabor and the Fisherfaces method to the reconstructed IR and visible images derived from wavelet frequency subbands. Our objective is to search for the subbands that are insensitive to the variation in expression and in illumination. The classification performance is improved by combining the multispectal information coming from the subbands that attain individually low equal error rate. Experimental results on Notre Dame face database show that the proposed wavelet-based algorithm outperforms previous multispectral images fusion method as well as monospectral method.

References

[1] ZhaoW., ChellappaR., PhillipsP., RosenfeldA.. ACM Computing Surveys, 2003, 35: 399

[2] SinghR., VatsaM., NooreA.. Pattern Recognition, 2008, 41: 880

[3] KongS.G., HeoJ., BoughorbelF., ZhengY., AbidiB.R., KoschanA., YiM., AbidiM.A.. International Journal of Computer Vision, 2007, 71: 215

[4] NanniL., LuminiA.. Pattern Recognition Letters, 2008, 29: 343

[5] EkenelH.K., SankurB.. Image and Vision Computing, 2005, 23: 469

[6] MaY., LiS.. Optoelectronics Letters, 2006, 2: 383

[7] BelhumeurP.N., HespanhaJ.P., KriegmanD.J.. IEEE Trans. Pattern Anal. Machine Intell., 2007, 19: 711

[8] LiuC., WechslerH.. IEEE Trans. Image Processing, 2002, 11: 467

[9] X. Chen, P.J. Flynn, and K.W. Bowyer, Proc. of ACM Workshop on Multimodal User Authentication, 2003, 48.

[10] MaioD., MaltoniD., JainA.K., PrabhakarS.. Handbook of Fingerprint Recognition, 2003, New York, Springer, 13