Abstract

An algorithm is described that removes the noise from images without causing blurring or other distortions of edges. The problem of noise removal is posed as a restoration of an uncorrupted image, given additive noise. The restoration problem is solved by using a new minimization strategy called mean-field annealing (MFA). An a priori statistical model of the image is chosen that drives the minimization toward solutions that are locally homogeneous. The strategy for MFA is derived, and the resulting algorithm is discussed. Applications of the algorithm to both synthetic images and real images are presented.

© 1989 Optical Society of America

Mean-field approximation minimizes relative entropy

Griff L. Bilbro, Wesley E. Snyder, and Reinhold C. Mann
J. Opt. Soc. Am. A 8(2) 290-294 (1991)

Integration of stereo vision and optical flow by using an energy-minimization approach

Nasser M. Nasrabadi, Sandra P. Clifford, and Yi Liu
J. Opt. Soc. Am. A 6(6) 900-907 (1989)

Mean-field annealing for phase unwrapping

Sebastiano Stramaglia, Luciano Guerriero, Guido Pasquariello, and Nicola Veneziani
Appl. Opt. 38(8) 1377-1383 (1999)

Image gathering and restoration: information and visual quality

Judith A. McCormick, Rachel Alter-Gartenberg, and Friedrich O. Huck
J. Opt. Soc. Am. A 6(7) 987-1005 (1989)

Iterative linear minimum mean-square-error image restoration from partially known blur

Vladimir Z. Mesarović, Nikolas P. Galatsanos, and Miles N. Wernick
J. Opt. Soc. Am. A 17(4) 711-723 (2000)