Color orthogonal local binary patterns combination for image region description

Visual content description is a key issue for machine-based image analysis and understanding. A good visual descriptor should be both discriminative enough and computationally efficient while possessing some properties of robustness to viewpoint changes and lighting condition variations. In this paper, we propose several new local descriptors based on color orthogonal local binary patterns combination (OLBPC) for image region description. The aim is to increase both discriminative power and photometric invariance properties of the original LBP while keeping its computational efficiency. The experiments in three different applications show that the proposed descriptors outperform the popular SIFT and CS-LBP, and get comparable or even slightly better performances than the state-of-the-art color SIFT descriptors. Meanwhile, they could provide complementary information to the color SIFT, because a fusion of these two kinds of descriptors is found to perform clearly better than either of the two separately. Moreover, the proposed descriptors are more computationally efficient than the color SIFT (about 2 times faster). KeywordsLocal descriptor; Region description; Orthogonal local binary patterns combination; Color LBP descriptor; CSLBP; SIFT; Image matching; Object recognition; Scene classification