Abstract
We study two approaches to fuzzy information fusion for face recognition involving aggregation of local and global face information and a wavelet decomposition approach. The former is concerned with a fuzzy information fusion involving both feature-based (e.g., eye, nose, and mouth) and template-based (global face) approaches to face recognition. The latter is associated with fuzzy information fusion based on four subimages (that is approximation, horizontal, vertical, and diagonal detailed images) using wavelet decomposition method. Making use of these two approaches, we design classifiers based on the well-known fisherface method. We demonstrate that the proposed method comes with better performance when compared with other template-based techniques and shows substantial insensitivity to large variation in light direction and facial expression. The fusion of the individual classifiers is realized by means of the Choquet fuzzy integral. The experimental results produced for FERET face database with 600 frontal face images corresponding to 200 subjects quantifies the performance of the classifier and contrast it with some other approaches existing in the literature.