Several methods have been proposed to describe face images in order to recognize them automatically. Local\nmethods based on spatial histograms of local patterns (or operators) are among the best-performing ones. In this\npaper, a new method that allows to obtain more robust histograms of local patterns by using a more discriminative\nspatial division strategy is proposed. Spatial histograms are obtained from regions clustered according to the\nsemantic pixel relations, making better use of the spatial information. Here, a simple rule is used, in which pixels in an\nimage patch are clustered by sorting their intensity values. By exploring the information entropy on image patches,\nthe number of sets on each of them is learned. Besides, Principal Component Analysis with a Whitening process is\napplied for the final feature vector dimension reduction, making the representation more compact and discriminative.\nThe proposed division strategy is invariant to monotonic grayscale changes, and shows to be particularly useful when\nthere are large expression variations on the faces. The method is evaluated on three widely used face recognition\ndatabases: AR, FERET and LFW, with the very popular LBP operator and some of its extensions. Experimental results\nshow that the proposal not only outperforms those methods that use the same local patterns with the traditional\ndivision, but also some of the best-performing state-of-the-art methods.
Loading....