Asian Journal of Computer Science and Technology (AJCST)
Several Face Images Recognition in Uncontrolled Illumination Variation ImagesAuthor : M.Janani and G.Thiruselvi
Volume 1 No.1 January-June 2012 pp 25-28
This paper to study and present a face recognition system under uncontrolled illumination images. It consists of pre-processing method, a hybrid Fourier feature extraction and a score fusion. In the pre-processing stage a face image is transformed in to an illumination-insensitive image, called an “integral normalized gradient image”. The hybrids Fourier feature are extracted from different Fourier domains in different frequency bandwidth. In this paper the multiple Face Models are generated by plural normalized face images that have different eye distances. Finally to combine the scores from multiple complementary classifiers, a log likelihood ratio-based score fusion scheme is applied. This system using the Face recognition Grand Challenge (FRGC) experimental protocols is evaluated. it shows the average verification rate on 2-D face images under various environmental variations such as illumination changes, expression changes, and time elapses.
Face recognition, Face recognition grand challenge, Feature extraction, preprocessing, score fusion