Enhanced performance metrics for blind image restoration
ICIC'05 Proceedings of the 2005 international conference on Advances in Intelligent Computing - Volume Part I
IEEE Transactions on Image Processing
Bayesian multichannel image restoration using compound Gauss-Markov random fields
IEEE Transactions on Image Processing
Hi-index | 0.00 |
An information of image quality and / or performance of image restoration methods can be expressed by evaluative criteria. The classical criteria are used for rather unrealistic image evaluation while the original (undegraded) image is available. More difficult case is in image restoration assessment when the original is unknown. Blind criteria serve as an image quality estimator based on the restored image. Criteria are tested and compared on the synthetic degraded data set (the "ideal" is known) to set out the most appropriate criterium for the real astronomical images. Results are discussed from various points of view of image degradation process including iffluence of the variable noise.