A robust regularized restoration algorithm based on Topkis-Veinott optimization method

Daan Zhu, M. Razaz, R. Lee

Research output: Contribution to conferenceOther


We present a robust iterative algorithm for restoration of blurred and noisy images. The restoration model is first regularised and then solved iteratively. An optimal regularisation parameter is estimated using the Topkis-Veinott gradient method. This restoration algorithm is compared with other competitive restoration algorithms in the literature. Experimental results are presented to demonstrate the comparative important features of our algorithm, namely, robustness in the presence of noise and high quality restoration results.
Original languageEnglish
Number of pages4
Publication statusPublished - 2004
Event17th International Conference on Pattern Recognition - Cambridge, United Kingdom
Duration: 23 Aug 200426 Aug 2004


Conference17th International Conference on Pattern Recognition
Abbreviated titleICPR-2004
Country/TerritoryUnited Kingdom

Cite this