Abstract

The noise sensitivity of interpolation and extrapolation matrices is investigated. For certain bandwidth and truncation parameters, the interpolation matrix is shown to yield results at a lower noise level than the input data. The input noise level, however, can be lowered by filtering the result. The noise level in the interpolated interval is shown to be lower near where the image is known. The extrapolation matrix is shown to be ill-conditioned, thus demonstrating severe sensitivity to input noise.

© 1982 Optical Society of America

Closed form bandlimited image extrapolation

David K. Smith and Robert J. Marks
Appl. Opt. 20(14) 2476-2483 (1981)

Extrapolation, interpolation, and identification of spots in Hartmann patterns

Yobani Mejía
Appl. Opt. 53(26) 6073-6082 (2014)

Superresolving image restoration using linear programming

R. Mammone and G. Eichmann
Appl. Opt. 21(3) 496-501 (1982)

Digital restoration of incoherent bandlimited images

Junji Maeda and Kazumi Murata
Appl. Opt. 21(12) 2199-2204 (1982)

Effects of sampling on closed form bandlimited signal interval interpolation

Robert J. Marks and Shiao-Min Tseng
Appl. Opt. 24(6) 763-765 (1985)