This article proposes a new no-reference image quality assessment method that is able to blindly predict the quality of an image. The method is based on a machine learning technique that uses texture descriptors. In the proposed method, texture features are computed by decomposing images into texture information using multiscale local binary pattern (MLBP) operators. In particular, the parameters of local binary pattern operators are varied, which generates MLBP operators. The features used for training the prediction algorithm are the histograms of these MLBP channels. The results show that, when compared with other state-of-the-art no-reference methods, the proposed method is competitive in terms of prediction precision and computational complexity. © 2016 Society for Imaging Science and Technology.
Texture, along with color, is one of the most important characteristics of a material defining its appearance. While color had been studied for a long time and continues being an interesting topic, the analysis of texture has traditionally been postponed, mainly because of its difficulty, and remains a challenge. Depending on the application, different approaches to texture characterization have been proposed in the bibliography. In this work, texture is considered in the context of visual perception and the second order statistical measurements based on the Grey-Level Co-occurrence Matrix (GLCM) have been computed for a database of texture images (KTH-TIPS and KTH-TIPS2). In the literature, there is no available information about the number of features needed for texture characterization, although no less than five parameters are typically employed. In our previous work, the selection of the optimal texture features was studied through Principal Component Analysis (PCA), using only those that are statistically significant describing the studied textures. In this work, the texture features obtained were analyzed from a perceptual point of view.