The paper presents an unsupervised procedure for change mapping from two very-high-resolution (VHR) SAR acquisitions. Following the approach proposed in [1], the method exploits the characteristics of a robust, parametric, high-order statistics change feature by extending it to a multiscale version, namely Multi-scale Kullback-Leibler (MKL) feature. Its effectiveness is experimentally confirmed on simulated and real VHR amplitude SAR image pairs. The thresholding algorithm is revisited, optimized and tested on true CosmoSkyMed (CSK) images. The experimental tests demonstrate the robustness and quality of the proposed change mapping method, which can be applied for damage assessment in disaster management systems.

Garzelli, A. (2022). Unsupervised VHR SAR change mapping. In Proc. IEEE IGARSS 2022 (pp.683-686).

Unsupervised VHR SAR change mapping

Garzelli, Andrea
2022

Abstract

The paper presents an unsupervised procedure for change mapping from two very-high-resolution (VHR) SAR acquisitions. Following the approach proposed in [1], the method exploits the characteristics of a robust, parametric, high-order statistics change feature by extending it to a multiscale version, namely Multi-scale Kullback-Leibler (MKL) feature. Its effectiveness is experimentally confirmed on simulated and real VHR amplitude SAR image pairs. The thresholding algorithm is revisited, optimized and tested on true CosmoSkyMed (CSK) images. The experimental tests demonstrate the robustness and quality of the proposed change mapping method, which can be applied for damage assessment in disaster management systems.
Garzelli, A. (2022). Unsupervised VHR SAR change mapping. In Proc. IEEE IGARSS 2022 (pp.683-686).
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11365/1213916