Publication Date:
1986
abstract:
Noise reduction filters were tested using NOAA satellite images. The best results, in terms of noise suppression and blur attenuation are obtained by two kinds of digital filtering techniques: weighted median filter with central pixel weight = 3 and decision threshold =12; and extremal filter with maximum number of extremal pixels =3. The other kinds of techniques show a nonuniform behavior, mainly in terms of noise pixel leaving. Both the filters were used in iterative mode: the weighted median filter needs fewer iterations to reach a stable condition than the extremal filter. However the mean computing time per iteration of the extremal filter is 30% less than the weighted filter time.
Iris type:
04.01 Contributo in Atti di convegno
List of contributors:
Sacco, VINCENZO MARIA; Baronti, Stefano; Carla', Roberto
Book title:
ESA Proceedings of the 1986 International Geoscience and Remote Sensing Symposium (IGARSS '86) on Remote Sensing: Today's Solutions for Tomorrow's Information Needs