Combining different regularization terms and various loss functions, we can have many variants of such linear models [ 2].
Since that, many different regularization terms have been proposed.
To enforce the sparsity of the phase errors three different regularization terms are proposed within the same framework.
We use the sum of squared differences in the L 2 -norm sense as the similarity measure and introduce four different regularization terms.
In [3] we studied the same optimal control formulation of the image registration problem with different regularization terms and showed the existence and uniqueness of this optimization problem using appropriate theorems.
So far, dozens of algorithms have been proposed based on optimization with different regularization terms [ 16, 17, 19], source support constraint [ 13, 15, 21– 24], multi-spectral image acquisition [ 7, 14, 16, 17, 25], adaptive discretization [ 22, 26], and those mixed.
In a related work [9], a 3-D wavelet based method was presented for deformable image registration where different similarity measure, different regularization term, and different types of wavelets were used.
In practice, several different terms and definitions have been developed for retailing in multiple channels.
One unigene may be assign into several different GO-terms.
Our method incorporates L 2 -norm sense similarity measures with several different curvature-driven diffusion-based regularization terms such as anisotropic diffusion, mean curvature motion (MCM), and affine invariant MCM.
