Local Binary Patterns (LBPs) have been highly used in texture classification\nfor their robustness, their ease of implementation and their low computational\ncost. Initially designed to deal with gray level images, several methods based\non them in the literature have been proposed for images having more than\none spectral band. To achieve it, whether assumption using color information\nor combining spectral band two by two was done. Those methods use micro\nstructures as texture features. In this paper, our goal was to design texture\nfeatures which are relevant to color and multicomponent texture analysis\nwithout any assumption. Based on methods designed for gray scale images,\nwe find the combination of micro and macro structures efficient for multispectral\ntexture analysis.
Loading....