Center-symmetric local binary pattern based on weighted neighbor contribution

A novel descriptor based on MFCS-LBP (Multi-resolution Fusion Center-Symmetric Local Binary Patterns) which introduces the contribution of sample's neighbors is proposed in this paper. First, the contribution of the sample's neighbors to the center point is considered and the Contribution Weighting Center-Symmetric Local Binary Pattern (CWCS-LBP) descriptor is constructed. Then, in view of the stability of the feature regions under different resolution, a Multi-resolution Fusion Center-Symmetric Local Binary Pattern(MFCS-LBP) descriptor is presented by combining the CWCS-LBP descriptor and the CS-LBP descriptor based on the multi-resolution fusion strategy. Compared with the similar algorithms，the experimental results demonstrate that the proposed algorithm can not only greatly reduce the description time but also improve the description performances in the presence of scale change, image rotation, viewpoint change and image blur.