Uncertainty-Aware Visualization for Analyzing Heterogeneous Wildfire Detections

Annie Preston, Maksim Gomov, Kwan-Liu Ma

Research output: Contribution to journalArticle

Abstract

There is growing interest in using data science techniques to characterize and predict natural disasters and extreme weather events. Such techniques merge noisy data gathered in the real world, from sources such as satellite detections, with algorithms that strongly depend on the noise, resolution, and uncertainty in these data. In this study, we present a visualization approach for interpolating multiresolution, uncertain satellite detections of wildfires into intuitive visual representations. We use extrinsic, intrinsic, coincident, and adjacent uncertainty representations as appropriate for understanding the information at each stage. To demonstrate our approach, we use our framework to tune two different algorithms for characterizing satellite detections of wildfires.

Original languageEnglish (US)
JournalIEEE Computer Graphics and Applications
DOIs
StatePublished - Jan 1 2019

Fingerprint

Visualization
Satellites
Disasters
Uncertainty

Keywords

  • Data visualization
  • geographic visualization
  • Interpolation
  • Meteorology
  • MODIS
  • Real-time systems
  • remote sensing
  • Satellites
  • Uncertainty
  • Uncertainty quantification

ASJC Scopus subject areas

  • Software
  • Computer Graphics and Computer-Aided Design

Cite this

Uncertainty-Aware Visualization for Analyzing Heterogeneous Wildfire Detections. / Preston, Annie; Gomov, Maksim; Ma, Kwan-Liu.

In: IEEE Computer Graphics and Applications, 01.01.2019.

Research output: Contribution to journalArticle

@article{727ac2e2b25149e5947221f431ccd39d,
title = "Uncertainty-Aware Visualization for Analyzing Heterogeneous Wildfire Detections",
abstract = "There is growing interest in using data science techniques to characterize and predict natural disasters and extreme weather events. Such techniques merge noisy data gathered in the real world, from sources such as satellite detections, with algorithms that strongly depend on the noise, resolution, and uncertainty in these data. In this study, we present a visualization approach for interpolating multiresolution, uncertain satellite detections of wildfires into intuitive visual representations. We use extrinsic, intrinsic, coincident, and adjacent uncertainty representations as appropriate for understanding the information at each stage. To demonstrate our approach, we use our framework to tune two different algorithms for characterizing satellite detections of wildfires.",
keywords = "Data visualization, geographic visualization, Interpolation, Meteorology, MODIS, Real-time systems, remote sensing, Satellites, Uncertainty, Uncertainty quantification",
author = "Annie Preston and Maksim Gomov and Kwan-Liu Ma",
year = "2019",
month = "1",
day = "1",
doi = "10.1109/MCG.2019.2918158",
language = "English (US)",
journal = "IEEE Computer Graphics and Applications",
issn = "0272-1716",
publisher = "IEEE Computer Society",

}

TY - JOUR

T1 - Uncertainty-Aware Visualization for Analyzing Heterogeneous Wildfire Detections

AU - Preston, Annie

AU - Gomov, Maksim

AU - Ma, Kwan-Liu

PY - 2019/1/1

Y1 - 2019/1/1

N2 - There is growing interest in using data science techniques to characterize and predict natural disasters and extreme weather events. Such techniques merge noisy data gathered in the real world, from sources such as satellite detections, with algorithms that strongly depend on the noise, resolution, and uncertainty in these data. In this study, we present a visualization approach for interpolating multiresolution, uncertain satellite detections of wildfires into intuitive visual representations. We use extrinsic, intrinsic, coincident, and adjacent uncertainty representations as appropriate for understanding the information at each stage. To demonstrate our approach, we use our framework to tune two different algorithms for characterizing satellite detections of wildfires.

AB - There is growing interest in using data science techniques to characterize and predict natural disasters and extreme weather events. Such techniques merge noisy data gathered in the real world, from sources such as satellite detections, with algorithms that strongly depend on the noise, resolution, and uncertainty in these data. In this study, we present a visualization approach for interpolating multiresolution, uncertain satellite detections of wildfires into intuitive visual representations. We use extrinsic, intrinsic, coincident, and adjacent uncertainty representations as appropriate for understanding the information at each stage. To demonstrate our approach, we use our framework to tune two different algorithms for characterizing satellite detections of wildfires.

KW - Data visualization

KW - geographic visualization

KW - Interpolation

KW - Meteorology

KW - MODIS

KW - Real-time systems

KW - remote sensing

KW - Satellites

KW - Uncertainty

KW - Uncertainty quantification

UR - http://www.scopus.com/inward/record.url?scp=85067042319&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067042319&partnerID=8YFLogxK

U2 - 10.1109/MCG.2019.2918158

DO - 10.1109/MCG.2019.2918158

M3 - Article

C2 - 31135352

AN - SCOPUS:85067042319

JO - IEEE Computer Graphics and Applications

JF - IEEE Computer Graphics and Applications

SN - 0272-1716

ER -