Adaptive algorithm for discontinuity preserving image restoration
K. Sunil Kumar, U.B. Desai · 1993 IEEE International Symposium on Circuits and Systems · 2002
Conventional adaptive algorithms for image restoration tend to blur the image by smoothing over discontinuities. To overcome this, a discontinuity preserving adaptive algorithm (DPAA) for image restoration is developed. The basic idea is to combine the continuous line field model with the usual linear finite impulse response (FIR) model used for image restoration. The constructed cost function is minimized using the gradient descent algorithm in adaptive setup. Computational cost and simulation results are given to aid comparison of DPAA and the conventional 2-D least mean square (LSD) algorithm.>