Title :
A technique for image restoration based on recursive processing and error correction
Author_Institution :
Dipt. di Elettrotecnica, Elettronica ed Inf., Trieste Univ., Italy
Abstract :
A technique for restoration of images corrupted by mixed noise is presented. The proposed approach aims at removing the different noise components by means of a recursive process that minimizes filtering errors. The method combines simplicity and effectiveness. No complicated tuning of system parameters is necessary. A very accurate restoration of data corrupted by mixed Gaussian and impulse noise can be obtained. Experimental results show that the proposed method can largely outperform other techniques in the literature
Keywords :
Gaussian noise; error correction; filtering theory; fuzzy systems; image restoration; impulse noise; effectiveness; error correction; filtering errors; image restoration; impulse noise; mixed Gaussian impulse noise; mixed noise; recursive processing; restoration of data; simplicity; tuning; Error correction; Filtering; Filters; Fuzzy sets; Gaussian noise; Image restoration; Noise generators; Noise level; Pulse generation; Smoothing methods;
Conference_Titel :
Instrumentation and Measurement Technology Conference, 2000. IMTC 2000. Proceedings of the 17th IEEE
Conference_Location :
Baltimore, MD
Print_ISBN :
0-7803-5890-2
DOI :
10.1109/IMTC.2000.848674