We introduce in this paper a novel Independent Gabor wavelet Features (IGF) method for face recognition. The IGF method derives first an augmented Gabor feature vector based upon the Gabor wavelet transformation of face images and using different orientation and scale local features. Independent Component Analysis (ICA) operates then on the Gabor feature vector subject to sensitivity analysis for the ICA transformation. Finally, the IGF method applies the Probabilistic Reasoning Model for classification by exploiting the independence properties between the feature components derived by the ICA. The feasibility of the new IGF method has been successfully tested on face recognition using 600 FERET frontal face images corresponding to 200 subjects whose facial expressions and lighting conditions may vary. © Springer-Verlag Berlin Heidelberg 2001.
CITATION STYLE
Liu, C., & Wechsler, H. (2001). Face recognition using independent gabor wavelet features. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2091 LNCS, pp. 20–25). Springer Verlag. https://doi.org/10.1007/3-540-45344-x_3
Mendeley helps you to discover research relevant for your work.