Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Change Detection Method for High Resolution Remote Sensing Images Using Random Forest

Feng, Wenqing ; Sui, Haigang ; Tu, Jihui ; Sun, Kaimin and Huang, Weiming LU (2017) In Cehui Xuebao/Acta Geodaetica et Cartographica Sinica 46(11). p.1880-1890
Abstract

Studies based on object-based image analysis (OBIA) representing the paradigm shift in remote sensing image change detection (CD) have achieved remarkable progress in the last decade. Their aim has been developing more intelligent interpretation analysis methods in the future. The prediction effect and performance stability of random forest (RF), as a new kind of machine learning algorithm, are better than many single predictors and integrated forecasting method. This paper presents a novel RF OBIA method for high resolution remote sensing image CD that makes full use of the advantages of RF and OBIA. Firstly, the entropy rate segmentation algorithm is used to segment the image for the purpose of measuring the homogeneity of... (More)

Studies based on object-based image analysis (OBIA) representing the paradigm shift in remote sensing image change detection (CD) have achieved remarkable progress in the last decade. Their aim has been developing more intelligent interpretation analysis methods in the future. The prediction effect and performance stability of random forest (RF), as a new kind of machine learning algorithm, are better than many single predictors and integrated forecasting method. This paper presents a novel RF OBIA method for high resolution remote sensing image CD that makes full use of the advantages of RF and OBIA. Firstly, the entropy rate segmentation algorithm is used to segment the image for the purpose of measuring the homogeneity of super-pixels. Then the optimal image segmentation result is obtained from the evaluation index of the optimal super-pixel number. Afterwards, the spectral features and Gabor features of each super-pixelareextracted and used as feature datasets for the training of RF model. On the basis of the initial pixel-level CD result, the changed and unchanged samples are automatically selected and used to build the classifier model in order to get the final object-level CD result. Experimental results on Quickbird, IKONOS and SPOT-5 multi-spectral images show that the proposed method out performs the compared methods in the accuracy of CD.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Change detection, Feature, Random forest, Segmentation, Super-pixel
in
Cehui Xuebao/Acta Geodaetica et Cartographica Sinica
volume
46
issue
11
pages
11 pages
publisher
Editorial Department of Acta Geodaetica et Cartographica Sinica
external identifiers
  • scopus:85037810517
ISSN
1001-1595
DOI
10.11947/j.AGCS.2017.20170074
language
English
LU publication?
yes
id
6529232e-6901-480c-a707-9c3a8dcd52de
date added to LUP
2018-01-04 11:12:31
date last changed
2022-04-01 21:42:41
@article{6529232e-6901-480c-a707-9c3a8dcd52de,
  abstract     = {{<p>Studies based on object-based image analysis (OBIA) representing the paradigm shift in remote sensing image change detection (CD) have achieved remarkable progress in the last decade. Their aim has been developing more intelligent interpretation analysis methods in the future. The prediction effect and performance stability of random forest (RF), as a new kind of machine learning algorithm, are better than many single predictors and integrated forecasting method. This paper presents a novel RF OBIA method for high resolution remote sensing image CD that makes full use of the advantages of RF and OBIA. Firstly, the entropy rate segmentation algorithm is used to segment the image for the purpose of measuring the homogeneity of super-pixels. Then the optimal image segmentation result is obtained from the evaluation index of the optimal super-pixel number. Afterwards, the spectral features and Gabor features of each super-pixelareextracted and used as feature datasets for the training of RF model. On the basis of the initial pixel-level CD result, the changed and unchanged samples are automatically selected and used to build the classifier model in order to get the final object-level CD result. Experimental results on Quickbird, IKONOS and SPOT-5 multi-spectral images show that the proposed method out performs the compared methods in the accuracy of CD.</p>}},
  author       = {{Feng, Wenqing and Sui, Haigang and Tu, Jihui and Sun, Kaimin and Huang, Weiming}},
  issn         = {{1001-1595}},
  keywords     = {{Change detection; Feature; Random forest; Segmentation; Super-pixel}},
  language     = {{eng}},
  month        = {{11}},
  number       = {{11}},
  pages        = {{1880--1890}},
  publisher    = {{Editorial Department of Acta Geodaetica et Cartographica Sinica}},
  series       = {{Cehui Xuebao/Acta Geodaetica et Cartographica Sinica}},
  title        = {{Change Detection Method for High Resolution Remote Sensing Images Using Random Forest}},
  url          = {{http://dx.doi.org/10.11947/j.AGCS.2017.20170074}},
  doi          = {{10.11947/j.AGCS.2017.20170074}},
  volume       = {{46}},
  year         = {{2017}},
}