Stochastic simulation of the sea surface as imaged by a remote sensing system is considered. The sensed image is modelled as a compound random field with K distributed amplitude and arbitrary correlation properties. Texture simulation is obtained by resorting to the infinite divisibility of the gamma distribution. Theoretical analysis shows that some limitations exist for the maximum correlation coefficient which can be prescribed to the generated field. First and second-order statistics of sample images have been measured and ergodic properties have been validated
Stochastic simulation of the sea surface as imaged by a remote sensing system is considered. The sensed image is modelled as a compound random field with K distributed amplitude and arbitrary correlation properties. Texture simulation is obtained by resorting to the infinite divisibility of the Gamma distribution. Theoretical analysis shows that some limitations exist for the maximum correlation coefficient which can be prescribed to the generated field. First and second-order statistics of sample images have been measured and ergodic properties have been validated.
Modelling and simulation of bidimensional processes for remote sensing applications
M. DI BISCEGLIE;GALDI C
1993-01-01
Abstract
Stochastic simulation of the sea surface as imaged by a remote sensing system is considered. The sensed image is modelled as a compound random field with K distributed amplitude and arbitrary correlation properties. Texture simulation is obtained by resorting to the infinite divisibility of the gamma distribution. Theoretical analysis shows that some limitations exist for the maximum correlation coefficient which can be prescribed to the generated field. First and second-order statistics of sample images have been measured and ergodic properties have been validatedI documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
