On the Influence of Illumination Quality in 2D Facial Recognition
Detecting automatically whether a facial image is greatly affected or not by the illumination conditions, allows us in some cases to discard those deteriorated images for further recognition tasks or, in other cases, to apply a preprocessing method only to the images that really need it. With this aim, our paper presents a study on the isolated influence of illumination quality of 2D images in facial recognition. First, a fuzzy inference system is designed to be as an objective and automatic method to evaluate the illumination quality of facial patterns. Then, we estimate the best recognition result for the same images using different image classification methods. By combining both the illumination quality with the corresponding recognition results for same face images, and computing the regression line of this set of patterns, we detect a nearly-linear regression trend between illumination quality and recognition rate for the images tested. This result can then be used as a quality measure of patterns in 2D facial recognition, and also for deciding whether it is worth or not using these facial images in recognition tasks.
KeywordsIllumination quality Facial variations 2D facial recognition Anthropometric feature points Fuzzy inference system
Unable to display preview. Download preview PDF.
- 5.Gonzalez, R.C., Woods, R.E.: Digital Image Processing, 3rd edn. Prentice Hall (2008)Google Scholar
- 6.Gross, R., Baker, S., Matthews, I., Kanade, T.: Face Recognition across Pose and Illumination. In: Li, S.Z., Jain, A.K. (eds.) Handbook of Face Recognition, pp. 193–216. Springer (2004)Google Scholar
- 7.Koenen, R.: MPEG-4 Overview, International Organisation for Standarisation, ISO/IEC JTC1/SC29/WG11 No. 4668 (2002), http://www.oipf.tv/docs/mpegif/overview.pdf
- 10.Phillips, P.J., et al.: Overview of the face recognition grand challenge. In: Hebert, M., Kriegman, D. (eds.) Proc. IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2005), pp. 947–954 (2005)Google Scholar
- 11.Rizo-Rodrıguez, D., Mendez-Vazquez, H., Garcıa-Reyes, E.: An Illumination Quality Measure for Face Recognition. In: Ercil, A. (ed.) Proc. 20th International Conference on Pattern Recognition (ICPR 2010), pp. 1477–1480. IEEE Press (2010)Google Scholar
- 13.Sugeno, M.: Industrial Applications of Fuzzy Control. Elsevier Publishing Company (1985)Google Scholar
- 16.Zou, X., Kittler, J., Messer, K.: Illumination Invariant Face Recognition: A Survey. In: Bowyer, K.W. (ed.) Proc. First IEEE Intl. Conf. on Biometrics: Theory, Applications, and Systems (BTAS 2007), pp. 1–8 (2007)Google Scholar