Unsupervised change detection in very high spatial resolution COSMO-Skymed SAR images

Nicola Acito; Salvatore Resta; Marco Diani; Giovanni Corsini; Alessandro Rossi

doi:10.1117/12.974663

Translator Disclaimer

21 November 2012 Unsupervised change detection in very high spatial resolution COSMO-Skymed SAR images

Nicola Acito, Salvatore Resta, Marco Diani, Giovanni Corsini, Alessandro Rossi

Author Affiliations +

Proceedings Volume 8536, SAR Image Analysis, Modeling, and Techniques XII; 853602 (2012) https://doi.org/10.1117/12.974663
Event: SPIE Remote Sensing, 2012, Edinburgh, United Kingdom

Abstract

In this work we propose two pixel-wise change detection techniques for unsupervised network infrastructure monitoring in SAR imagery applications. The first algorithm is inspired by a well known algorithm, named RX, proposed to deal with anomaly detection in optical images. The second algorithm is a statistical based procedure, which exploits a nonparametric approach for estimating the probability density function of the image pair. In order to test and validate the proposed methods, we analyze a spot light amplitude COSMO-SkyMed image pair at one-meter spatial resolution acquired on a complex urban scenario. Experimental results obtained on the available dataset are presented and discussed.

Citation Download Citation

Nicola Acito, Salvatore Resta, Marco Diani, Giovanni Corsini, and Alessandro Rossi "Unsupervised change detection in very high spatial resolution COSMO-Skymed SAR images", Proc. SPIE 8536, SAR Image Analysis, Modeling, and Techniques XII, 853602 (21 November 2012); https://doi.org/10.1117/12.974663

ACCESS THE FULL ARTICLE