Nonparametric methods for detecting change in Multitemporal SAR/PolSAR Satellite Data

2020/01/16 by Fonseca, Rodney, Ludwig, Guilherme, Montoril, Michel +1
#62G07 #62M10 #Applications (stat.AP) #FOS: Computer and information sciences #G.3 #I.4.9

paper · doi:10.48550/arxiv.2001.05764

Abstract

We employ nonparametric statistical procedures to analyse multitemporal SAR/PolSAR satellite images. The aim is two-fold. We seek parsimony in data representation as well as efficient change detection. For these, wavelets and geostatistical analyses are applied to the images (Morettin et al., 2017; Krainski et al., 2018). Following this representation, the dimension of the underlying generating process is estimated (Fonseca and Pinheiro, 2019), and a set of multivariate characteristics is extracted. Change-points are then detected via wavelets (Montoril et al., 2019).

Related

• Wavelet Spatio-Temporal Change Detection on multi-temporal PolSAR images
• Change detection in SAR time-series based on the coefficient of variation
• Multitemporal SAR images change detection and visualization using RABASAR and simplified GLR
• Adaptive Block-Based Change-Point Detection for Sparse Spatially Clustered Data with Applications in Remote Sensing Imaging
• Online Change Detection in SAR Time-Series with Kronecker Product Structured Scaled Gaussian Models