Research Article
Statistical Texture Modeling for Medical Volume Using Linear Tensor Coding
Algorithm 2
Iteration algorithm of LTC.
IN: a series of th order tensors, , . Define Matrices | (, ) with orthogonal column vectors. | OUT: Rank-1 basis tensor , depends on convergence. | Iterate for until convergence | (1) Initial values: , define . | (2) (a) Initial values: and whose columns are determined as the first leading | eigenvectors of the matrices . | (b) Iterate for until convergence | (i) Maximize , | Solution: whose columns are determined as the first leading eigenvectors | of | Set . | (ii) Maximize , | Solution: whose columns are determined as the first leading eigenvectors | of | Set . | | (iii) Maximize , | | Solution: whose columns are determined as the first leading eigenvectors | of | Set . | | (iv) Maximize , | Solution: whose columns are determined as the first leading eigenvectors | of | Set . | | (c) Set , . | (3) Size of is , each basis . | (4) For each data, coefficient on this basis . | (5) For each data . |
|