LGSLRR: Towards fusing discriminative ordinal local and global structured low-rank representation for image recognition

摘要: Abstract Structural information extraction has been a focal technique in many classification applications, such as image recognition and biometrics. However, it remains challenge to simultaneously utilize local global structural model. In addition, terms of the information, existing methods mainly seek extract or preserve first-order structure while ignoring useful ordinal for classification. To this end, paper presents discriminative structured low-rank representation (LGSLRR) model that jointly preserves recognition. A block-diagonal is employed obtain second-order preserved by joint graph based manifold embedding with two different Laplacian matrices. Some extensive comparison experiments on ten public datasets are performed, results demonstrate effectiveness significant performance proposed method over some state-of-the-art methods.