Using independent subspace analysis for selecting filters used in texture processing

In this work we propose the use of independent subspace analysis (ISA) for selecting filters used for texture processing. ISA is an extension of independent component analysis (ICA), a technique employed to decompose an image into statistically independent features. In ISA, complete independence of features is not required; features that possess some mutual dependence are associated in feature subspaces. An emergent characteristic of the ISA model is that these subspaces enclose features of similar frequency content and orientation. This, in turn, helps in determining a reduced set of filters later employed in a texture classification task. Preliminary results here presented show that our proposed ISA criterion can attain performance comparable to other filter based classification schemes while resulting in a considerably smaller filter bank.