Restoration of bandlimited images by an iterative damped least-squares method with adaptive regularization

Publisher Website
Google Scholar
Abstract
A digital method for restoring bandlimited images in the presence of noise is described. The restoration procedure is an iterative damped least-squares algorithm that is based on the principle of damped least squares. This method has the advantage of tolerance to noise by incorporating additional constraints of non-negativity of the object and the adaptive regularization as well as a priori information about the finite extent of the object. We present some results of computer simulations of both 1-D and 2-D images that demonstrate the effectiveness of the proposed method.