We present a novel approach for face recognition by combining a local binary pattern (LBP)-based face descriptor and the distinctive information of faces. Several studies of psychophysics have shown that the eyes or mouth can be an important cue in human face perception, and the nose plays an insignificant role. This means that there exists a distinctive information distribution of faces. First, we give a quantitative estimation of the density for each pixel in a fronted face image by combining the Parzen-window approach and scale invariant feature transform detector, which is taken as the measure of the distinctive information of the faces. Second, we integrate the density function in the subwindow region of the face to gain the weight set used in the LBP-based face descriptor to produce weighted chi-square statistics. As an elementary application of the estimation of distinctive information of faces, the proposed method is tested on the FERET FA/FB image sets and yields a recognition rate of 98.2% compared to the 97.3% produced by the method adopted by Ahonen, Hadid, and Pietikainen.
In this paper, we propose an Infrared(IR) background simulation method by integrating texture modeling and
infrared prediction. First, by introducing the latest work of compute vision, we argue that the infrared texture
play more important role of scene configuration contrast to the traditional viewpoint that the infrared texture is
just used to overcome low resolution of model or for feature enhancement. Next, we present the infrared texture
should be simulated in the radiant energy space relative to the temperature field, and synthesize the infrared
texture using FRAME model. In the end, according to the scene model introduced from the compute vision
context, we present the IR background simulation method by integrating the IR prediction of material region
and the corresponding texture synthesized by FRAME model.