Steerable Pyramids Feature Based Classification Using Fisher Linear Discriminant for Face Recognition

In this paper, an efficient local appearance feature extraction method based the multiresolutionSteerable Pyramids (SP) transform is proposed in order to further enhance the performance ofthe well known Fisher Linear Discriminant (FLD) method when applied to face recognition. Each face isdescribed by a subset of band filtered images containing block-based SP coefficients. These coefficientscharacterize the face texture and a set of simple statistical measures allows us to form compact and meaningfulfeature vectors. The proposed method is compared with some related feature extraction methodssuch as Principal component analysis (PCA), as well as Linear Discriminant Analysis, and Fisher LinearDiscriminant (FLD), Independent Component Analysis and ICA. Experimental results on ORL, YALEand FERET face databases convince us that the proposed method provides a better representation of theclass information and obtains much higher recognition accuracies