Noisy image restoration by cost function minimization

Implementation of a new approach to the nonlinear noisy image restoration problem, which is feasible for large images, is described. The approach is based on a vector space representation of images and on using the conjugate gradient algorithm to solve a least squares minimization problem. Computer simulations yield good results with relatively little computational cost.

[1]  Shmuel Peleg,et al.  PICTURES AS ELEMENTS IN VECTOR SPACE. , 1983, CVPR 1983.

[2]  William K. Pratt,et al.  Fast Computational Techniques for Pseudoinverse and Wiener Image Restoration , 1977, IEEE Transactions on Computers.

[3]  John Skilling,et al.  Image restoration by a powerful maximum entropy method , 1982, Comput. Vis. Graph. Image Process..

[4]  Azriel Rosenfeld,et al.  Digital Picture Processing , 1976 .

[5]  Shmuel Peleg,et al.  Inversion of picture operators , 1987, Pattern Recognit. Lett..