Abstract
We describe a new digital method of superresolution or restoration of band-limited images in the presence of noise. The restoration procedure is an iterative regularized pseudoinverse (RPI) algorithm that is based on the principle of least squares. This method acquires the advantage of tolerance to noise by incorporating additional constraints of nonnegativity of the object and adaptive regularization as well as the finite extent of the object. After discussing the convergence of the iterative RPI algorithm, we present some results of computer simulations that demonstrate the effectiveness of the proposed method.
© 1984 Optical Society of America
Full Article | PDF ArticleMore Like This
Junji Maeda and Kazumi Murata
Appl. Opt. 23(6) 857-861 (1984)
Junji Maeda
Appl. Opt. 24(10) 1421-1425 (1985)
Junji Maeda
Appl. Opt. 24(6) 751-757 (1985)