Comparison of remote sensing image processing techniques to identify tornado damage areas from Landsat TM data

Soe Myint, May Yuan, Randall Cerveny, Chandra P. Giri

Research output: Contribution to journalArticle

36 Citations (Scopus)

Abstract

Remote sensing techniques have been shown effective for large-scale damage surveys after a hazardous event in both near real-time or post-event analyses. The paper aims to compare accuracy of common imaging processing techniques to detect tornado damage tracks from Landsat TM data. We employed the direct change detection approach using two sets of images acquired before and after the tornado event to produce a principal component composite images and a set of image difference bands. Techniques in the comparison include supervised classification, unsupervised classification, and objectoriented classification approach with a nearest neighbor classifier. Accuracy assessment is based on Kappa coefficient calculated from error matrices which cross tabulate correctly identified cells on the TM image and commission and omission errors in the result. Overall, the Object-oriented Approach exhibits the highest degree of accuracy in tornado damage detection. PCA and Image Differencing methods show comparable outcomes. While selected PCs can improve detection accuracy 5 to 10%, the Object-oriented Approach performs significantly better with 15-20% higher accuracy than the other two techniques.

Original languageEnglish (US)
Pages (from-to)1128-1156
Number of pages29
JournalSensors
Volume8
Issue number2
StatePublished - Feb 2008

Fingerprint

Tornadoes
tornadoes
image processing
remote sensing
Remote sensing
Image processing
damage
Passive Cutaneous Anaphylaxis
Damage detection
Classifiers
change detection
Imaging techniques
classifiers
Composite materials
Processing
composite materials
coefficients
matrices
cells

Keywords

  • Change detection
  • Damage
  • Image differencing
  • Object-oriented
  • Principal component

ASJC Scopus subject areas

  • Analytical Chemistry
  • Biochemistry

Cite this

Comparison of remote sensing image processing techniques to identify tornado damage areas from Landsat TM data. / Myint, Soe; Yuan, May; Cerveny, Randall; Giri, Chandra P.

In: Sensors, Vol. 8, No. 2, 02.2008, p. 1128-1156.

Research output: Contribution to journalArticle

@article{2ca540ec825d4f77867045b43fcabfd7,
title = "Comparison of remote sensing image processing techniques to identify tornado damage areas from Landsat TM data",
abstract = "Remote sensing techniques have been shown effective for large-scale damage surveys after a hazardous event in both near real-time or post-event analyses. The paper aims to compare accuracy of common imaging processing techniques to detect tornado damage tracks from Landsat TM data. We employed the direct change detection approach using two sets of images acquired before and after the tornado event to produce a principal component composite images and a set of image difference bands. Techniques in the comparison include supervised classification, unsupervised classification, and objectoriented classification approach with a nearest neighbor classifier. Accuracy assessment is based on Kappa coefficient calculated from error matrices which cross tabulate correctly identified cells on the TM image and commission and omission errors in the result. Overall, the Object-oriented Approach exhibits the highest degree of accuracy in tornado damage detection. PCA and Image Differencing methods show comparable outcomes. While selected PCs can improve detection accuracy 5 to 10{\%}, the Object-oriented Approach performs significantly better with 15-20{\%} higher accuracy than the other two techniques.",
keywords = "Change detection, Damage, Image differencing, Object-oriented, Principal component",
author = "Soe Myint and May Yuan and Randall Cerveny and Giri, {Chandra P.}",
year = "2008",
month = "2",
language = "English (US)",
volume = "8",
pages = "1128--1156",
journal = "Sensors (Switzerland)",
issn = "1424-8220",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "2",

}

TY - JOUR

T1 - Comparison of remote sensing image processing techniques to identify tornado damage areas from Landsat TM data

AU - Myint, Soe

AU - Yuan, May

AU - Cerveny, Randall

AU - Giri, Chandra P.

PY - 2008/2

Y1 - 2008/2

N2 - Remote sensing techniques have been shown effective for large-scale damage surveys after a hazardous event in both near real-time or post-event analyses. The paper aims to compare accuracy of common imaging processing techniques to detect tornado damage tracks from Landsat TM data. We employed the direct change detection approach using two sets of images acquired before and after the tornado event to produce a principal component composite images and a set of image difference bands. Techniques in the comparison include supervised classification, unsupervised classification, and objectoriented classification approach with a nearest neighbor classifier. Accuracy assessment is based on Kappa coefficient calculated from error matrices which cross tabulate correctly identified cells on the TM image and commission and omission errors in the result. Overall, the Object-oriented Approach exhibits the highest degree of accuracy in tornado damage detection. PCA and Image Differencing methods show comparable outcomes. While selected PCs can improve detection accuracy 5 to 10%, the Object-oriented Approach performs significantly better with 15-20% higher accuracy than the other two techniques.

AB - Remote sensing techniques have been shown effective for large-scale damage surveys after a hazardous event in both near real-time or post-event analyses. The paper aims to compare accuracy of common imaging processing techniques to detect tornado damage tracks from Landsat TM data. We employed the direct change detection approach using two sets of images acquired before and after the tornado event to produce a principal component composite images and a set of image difference bands. Techniques in the comparison include supervised classification, unsupervised classification, and objectoriented classification approach with a nearest neighbor classifier. Accuracy assessment is based on Kappa coefficient calculated from error matrices which cross tabulate correctly identified cells on the TM image and commission and omission errors in the result. Overall, the Object-oriented Approach exhibits the highest degree of accuracy in tornado damage detection. PCA and Image Differencing methods show comparable outcomes. While selected PCs can improve detection accuracy 5 to 10%, the Object-oriented Approach performs significantly better with 15-20% higher accuracy than the other two techniques.

KW - Change detection

KW - Damage

KW - Image differencing

KW - Object-oriented

KW - Principal component

UR - http://www.scopus.com/inward/record.url?scp=40849096430&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=40849096430&partnerID=8YFLogxK

M3 - Article

VL - 8

SP - 1128

EP - 1156

JO - Sensors (Switzerland)

JF - Sensors (Switzerland)

SN - 1424-8220

IS - 2

ER -