K-Bessel regression model for speckled data

A. D. C. Nascimento, P. M. Almeida-Junior, J. M. Vasconcelos and A. P. M. Borges-Junior

Journal of Applied Statistics, 2024, vol. 51, issue 2, 324-347

Abstract: Synthetic aperture radar (SAR) provides an efficient way to monitor the Earth's surface. But the speckle noise that the SAR system generates when acquiring images makes it difficult to understand and interpret SAR intensity features. To automatically analyze SAR images, this paper presents a K-Bessel regression (KBR) model in which a function of the mean intensity response is explained by other features (or covariates) determined in parallel. Some mathematical properties of this regression are derived and discussed in the context of the physical origin of the SAR image. A maximum likelihood estimation procedure is planned and its performance is quantified by Monte Carlo experiments. An application to real data obtained from a polarimetric SAR image of San Francisco Bay is realized. Results show that both the KBR-based processing is more informative than the unconditional approach to describe SAR intensity and that our proposal can outperform the normal and gamma regression models. Finally, it is shown that the KBR model is useful to reproduce the relief signal of one channel from the intensity values of the other.

Date: 2024
References: Add references at CitEc
Citations:

Downloads: (external link)
http://hdl.handle.net/10.1080/02664763.2022.2125935 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:taf:japsta:v:51:y:2024:i:2:p:324-347

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/CJAS20

DOI: 10.1080/02664763.2022.2125935

Access Statistics for this article

Journal of Applied Statistics is currently edited by Robert Aykroyd

More articles in Journal of Applied Statistics from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().