Visible light texture image classification using Gabor and LBP feature

LBP and Gabor wavelets are two widely used and successful local image representation methods. This paper presents the visible light texture image classification method simultaneously using Gabor and Local Binary Patterns (LBP) features. We design two kinds of feature fusion methods using the two types of the representations, which are performed in feature level and matching score level, respectively, and successfully applied them to visible light texture image classification. The proposed method is reasonable, because the Gabor and LBP features provide complementary description of the visible light texture image. Experiment results on MIT texture database demonstrate the effectiveness of our method. © 2013 Binary Information Press.