A comprehensive benchmark of local binary pattern algorithms for texture retrieval

Image retrieval is a well researched area and often based on integrating various kinds of image features. Apart from colour features, texture features are deemed crucial for successful image retrieval. Local binary pattern (LBP) based texture algorithms have gained significant popularity in recent years and have been shown to be useful for a variety of tasks. In this paper, we provide a comprehensive benchmark of LBP based methods for texture retrieval. In particular, a comparison of 16 LBP variants leading to 38 different texture descriptors, are evaluated on a large dataset of more than 6000 texture images. Interestingly, conventional LBP features are shown to work best, while almost all LBP methods are shown to significantly outperform other texture methods including Tamura, co-occurrence and Gabor features.