It is reasonable to combine them together to achieve impressive classification performance, as the combined feature takes the advantages of the two kinds of features in different aspects. Finally, the authors design a Co-occurrence Differential Texton (CoDT) feature to represent the local image patches of HEp-2 cells.