Title :
Information extraction form single look complex SAR images
Author :
Dusšan Gleich;Žarko Čučej
Author_Institution :
University of Maribor, Faculty of Electrical Engineering and Computer Science, Smetanova 17, 2000, Slovenia
fDate :
7/1/2011 12:00:00 AM
Abstract :
This paper proposes despeckling using the regularization framework. In order to separate the scene and speckle an auto-binomial model is used. The cost function consists of an auto-binomial prior, system response function and the differential part. The minimization problem is solved using a quasi-Newton approach. The system function term estimates the amount of speckle, the prior estimates the properties of the scene, the differential is used to preserve the point like characteristics of SAR image.
Keywords :
"Speckle","Approximation methods","Approximation algorithms","Cost function","Computational modeling","Data mining","Bayesian methods"
Conference_Titel :
Geoscience and Remote Sensing Symposium (IGARSS), 2011 IEEE International
Print_ISBN :
978-1-4577-1003-2
Electronic_ISBN :
2153-7003
DOI :
10.1109/IGARSS.2011.6050028