Kopriva, Ivica; Cichocki, Andrzej
Blind Multi-spectral Image Decomposition by 3D Nonnegative Tensor Factorization.
Optics Letters, 34
Alpha-divergence based nonnegative tensor factorization (NTF) is applied to blind multi-spectral image (MSI) decomposition. Matrix of spectral profiles and matrix of spatial distributions of the materials resident in the image are identified from the factors in Tucker3 and PARAFAC models. NTF preserves local structure in the MSI that is lost, due to vectorization of the image, with nonnegative matrix factorization (NMF)- or independent component analysis (ICA)-based decompositions. Moreover, NTF based on PARAFAC model is unique up to permutation and scale under mild conditions. To achieve this, NMF- and ICA-based factorizations respectively require enforcement of sparseness (orthogonality) and statistical independence constraints on the spatial distributions of the materials resident in the MSI, and that is not true. We demonstrate efficiency of the NTF-based factorization in relation to NMF- and ICA-based factorizations on blind decomposition of the experimental MSI with the known ground truth.
Actions (login required)