Local-statistics speckle filtering has been extended to multi-temporal SAR data by exploiting the space-varying temporal correlation of speckle noise between two images of the same scene taken at different times. A recursive nonlinear transformation aimed at decorrelating the data across time while retaining the multiplicative noise model is defined from the pixel geometric mean and ratio of a couple of spatially overlapped observations. The average temporal correlation coefficient is estimated from the scatterplots of local standard deviation to local mean calculated on transformed couples of images, through an unsupervised clustering procedure. The images are filtered in the transformed domain and reversely transformed to yield despeckled observations in which seasonal changes are preserved, or even highlighted, and texture analysis is expedited. Tests on two SAR images from repeat-pass ERS-1 corroborate the theoretical assumptions and show the performances of the proposed approach.
Joint change analysis and speckle filtering of multitemporal ERS-1 imagery / Alparone, Luciano; Baronti, Stefano; Garzelli, Andrea. - STAMPA. - 3:(1999), pp. 1543-1545. (Intervento presentato al convegno Proceedings of the 1999 IEEE International Geoscience and Remote Sensing Symposium (IGARSS'99) 'Remote Sensing of the Systems Earth - A Challenge for the 21st Century' tenutosi a Hamburg, Ger nel 28 June - 2w July 1999) [10.1109/IGARSS.1999.772013].
Joint change analysis and speckle filtering of multitemporal ERS-1 imagery
ALPARONE, LUCIANO;
1999
Abstract
Local-statistics speckle filtering has been extended to multi-temporal SAR data by exploiting the space-varying temporal correlation of speckle noise between two images of the same scene taken at different times. A recursive nonlinear transformation aimed at decorrelating the data across time while retaining the multiplicative noise model is defined from the pixel geometric mean and ratio of a couple of spatially overlapped observations. The average temporal correlation coefficient is estimated from the scatterplots of local standard deviation to local mean calculated on transformed couples of images, through an unsupervised clustering procedure. The images are filtered in the transformed domain and reversely transformed to yield despeckled observations in which seasonal changes are preserved, or even highlighted, and texture analysis is expedited. Tests on two SAR images from repeat-pass ERS-1 corroborate the theoretical assumptions and show the performances of the proposed approach.
I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
