Quantifying the Influence of Intensity Channels from Polsar Images for Edge Detection on Information Fusion
Authors:
- 11 July 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1514-1517
Abstract
A prompt and effective response to natural disasters is essential. Polarimetric Synthetic Aperture Radar (PolSAR) images playa crucial role in helping decisions in these scenarios, especially in adverse weather or the absence of sunlight. In this context, PolSAR images still provide information. In these situations, the process of finding precise boundaries to determine the extent of the disaster is an essential issue as PolSAR image resolutions are usually high (of the order of 10 m). The fusion of information from the intensity channels of PolSAR images for edge detection enhances the precision of edges presented in these images. In this work, we propose three metrics: the accuracy (Mac), the $\mathrm{F}_{1}$ -score (Mfe), and the Matthews correlation coefficient normalized (nMcc). These metrics quantify the influence of the information from the intensity channels of PolSAR images on the information fusion process for edge detection.
Keywords
This publication has 12 references indexed in Scilit:
- Fusion of Evidences in Intensity Channels for Edge Detection in PolSAR ImagesIEEE Geoscience and Remote Sensing Letters, 2020
- A Badging System for Reproducibility and Replicability in Remote Sensing ResearchIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020
- The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluationBMC Genomics, 2020
- A Survey of Predictive Modeling on Imbalanced DomainsACM Computing Surveys, 2016
- Fusion of polarimetric and texture information for urban building extraction from fully polarimetric SAR imageryRemote Sensing Letters, 2015
- Comparing Edge Detection Methods Based on Stochastic Entropies and Distances for PolSAR ImageryIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2013
- Generalized Simulated Annealing for Global Optimization: The GenSA PackageThe R Journal, 2013
- maxLik: A package for maximum likelihood estimation in RComputational Statistics, 2010
- Polarimetric SAR image segmentation with B-splines and a new statistical modelMultidimensional Systems and Signal Processing, 2010
- Feature extraction in speckled imagery using dynamic B‐spline deformable contours under the modelInternational Journal of Remote Sensing, 2006