Abstract:
|
This paper addresses the problem of interferometric
phase noise reduction in synthetic aperture radar interferometry.
A new phase noise model in the complex domain is introduced and
validated by using both simulated and real interferograms. This
noise model is also derived in the complex wavelet domain, where
a novel noise reduction algorithm, which is not based on a windowing
process and without the necessity of phase unwrapping,
is addressed. The use of the wavelet transform allows to maintain
the spatial resolution in the filtered phase image and prevents to
filter low coherence areas. By using both, simulated as well as real
interferometric phase images, the performance of this algorithm,
in terms of noise reduction, spatial resolution maintenance, and
computational efficiency, is reported and compared with other conventional
filtering approaches. |