“…The a priori knowledge gives a way to moderately reduce the demand for input data. Usually the a priori information is introduced in form of functionals [3,[7][8][9][12][13][14] that tend to support, in weighted form, some important expected features of the reconstructed image, such as smoothness [15,16], minimal summarized grey level in the image [7], monotonicity, convexity, porosity [17], etc. Much power is enclosed in the statistical methods for reconstruction [18][19][20] that support the maximum entropy, minimal Gibbs energy or other probabilistic priors in the image under restoration.…”