Study on extraction methods of ocean surface oil spill using HJ-CCD data

Yingying Gai, Bin Zhou, Yuanfang Sun, Yan Zhou, WenWen Li

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Oil spill is one of the most common ocean environmental disasters. Rapid and accurate access to the oil spill information has great significances to dynamic monitoring, conservation and sustainable use of the ocean. HJ-1 is a new satellite platform designed for environmental disasters. However, multi-spectral images obtained from HJ-1 have fewer spectral bands and the accuracy of acquiring the oil spill coverage only by spectral information is low. In this paper, oil spill occurred in Gulf of Mexico was selected as the research object. The texture characteristics which affect oil spill identification were extracted by Gray Level Co-occurrence Matrix (GLCM) and a decision tree method combined spectral characteristics with texture characteristics was established to extract the oil spill. And comparative analysis with the extraction result of Maximum Likelihood Classification (MLC) was performed. Results showed that, compared with MLC, the decision tree classification raised the user accuracy and the producer accuracy of extracting oil spill by 11.85% and 4.28%, respectively.

Original languageEnglish (US)
Title of host publicationInternational Conference on Remote Sensing, Environment and Transportation Engineering, RSETE 2013
PublisherAtlantis Press
Pages756-759
Number of pages4
ISBN (Print)9789078677772
StatePublished - 2013
Event2013 International Conference on Remote Sensing, Environment and Transportation Engineering, RSETE 2013 - Nanjing, China
Duration: Jul 26 2013Jul 28 2013

Other

Other2013 International Conference on Remote Sensing, Environment and Transportation Engineering, RSETE 2013
CountryChina
CityNanjing
Period7/26/137/28/13

Fingerprint

Oil spills
Charge coupled devices
disaster
Decision trees
producer
Disasters
Maximum likelihood
Mexico
conservation
coverage
monitoring
Textures
Conservation
Satellites
Monitoring

Keywords

  • Decision tree
  • Gulf of Mexico
  • HJ-1
  • Oil spill
  • Texture characteristic

ASJC Scopus subject areas

  • Computer Science Applications
  • Transportation

Cite this

Gai, Y., Zhou, B., Sun, Y., Zhou, Y., & Li, W. (2013). Study on extraction methods of ocean surface oil spill using HJ-CCD data. In International Conference on Remote Sensing, Environment and Transportation Engineering, RSETE 2013 (pp. 756-759). Atlantis Press.

Study on extraction methods of ocean surface oil spill using HJ-CCD data. / Gai, Yingying; Zhou, Bin; Sun, Yuanfang; Zhou, Yan; Li, WenWen.

International Conference on Remote Sensing, Environment and Transportation Engineering, RSETE 2013. Atlantis Press, 2013. p. 756-759.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Gai, Y, Zhou, B, Sun, Y, Zhou, Y & Li, W 2013, Study on extraction methods of ocean surface oil spill using HJ-CCD data. in International Conference on Remote Sensing, Environment and Transportation Engineering, RSETE 2013. Atlantis Press, pp. 756-759, 2013 International Conference on Remote Sensing, Environment and Transportation Engineering, RSETE 2013, Nanjing, China, 7/26/13.
Gai Y, Zhou B, Sun Y, Zhou Y, Li W. Study on extraction methods of ocean surface oil spill using HJ-CCD data. In International Conference on Remote Sensing, Environment and Transportation Engineering, RSETE 2013. Atlantis Press. 2013. p. 756-759
Gai, Yingying ; Zhou, Bin ; Sun, Yuanfang ; Zhou, Yan ; Li, WenWen. / Study on extraction methods of ocean surface oil spill using HJ-CCD data. International Conference on Remote Sensing, Environment and Transportation Engineering, RSETE 2013. Atlantis Press, 2013. pp. 756-759
@inproceedings{c2a6b3a5e2d54c31b78ff218e25ffbf4,
title = "Study on extraction methods of ocean surface oil spill using HJ-CCD data",
abstract = "Oil spill is one of the most common ocean environmental disasters. Rapid and accurate access to the oil spill information has great significances to dynamic monitoring, conservation and sustainable use of the ocean. HJ-1 is a new satellite platform designed for environmental disasters. However, multi-spectral images obtained from HJ-1 have fewer spectral bands and the accuracy of acquiring the oil spill coverage only by spectral information is low. In this paper, oil spill occurred in Gulf of Mexico was selected as the research object. The texture characteristics which affect oil spill identification were extracted by Gray Level Co-occurrence Matrix (GLCM) and a decision tree method combined spectral characteristics with texture characteristics was established to extract the oil spill. And comparative analysis with the extraction result of Maximum Likelihood Classification (MLC) was performed. Results showed that, compared with MLC, the decision tree classification raised the user accuracy and the producer accuracy of extracting oil spill by 11.85{\%} and 4.28{\%}, respectively.",
keywords = "Decision tree, Gulf of Mexico, HJ-1, Oil spill, Texture characteristic",
author = "Yingying Gai and Bin Zhou and Yuanfang Sun and Yan Zhou and WenWen Li",
year = "2013",
language = "English (US)",
isbn = "9789078677772",
pages = "756--759",
booktitle = "International Conference on Remote Sensing, Environment and Transportation Engineering, RSETE 2013",
publisher = "Atlantis Press",

}

TY - GEN

T1 - Study on extraction methods of ocean surface oil spill using HJ-CCD data

AU - Gai, Yingying

AU - Zhou, Bin

AU - Sun, Yuanfang

AU - Zhou, Yan

AU - Li, WenWen

PY - 2013

Y1 - 2013

N2 - Oil spill is one of the most common ocean environmental disasters. Rapid and accurate access to the oil spill information has great significances to dynamic monitoring, conservation and sustainable use of the ocean. HJ-1 is a new satellite platform designed for environmental disasters. However, multi-spectral images obtained from HJ-1 have fewer spectral bands and the accuracy of acquiring the oil spill coverage only by spectral information is low. In this paper, oil spill occurred in Gulf of Mexico was selected as the research object. The texture characteristics which affect oil spill identification were extracted by Gray Level Co-occurrence Matrix (GLCM) and a decision tree method combined spectral characteristics with texture characteristics was established to extract the oil spill. And comparative analysis with the extraction result of Maximum Likelihood Classification (MLC) was performed. Results showed that, compared with MLC, the decision tree classification raised the user accuracy and the producer accuracy of extracting oil spill by 11.85% and 4.28%, respectively.

AB - Oil spill is one of the most common ocean environmental disasters. Rapid and accurate access to the oil spill information has great significances to dynamic monitoring, conservation and sustainable use of the ocean. HJ-1 is a new satellite platform designed for environmental disasters. However, multi-spectral images obtained from HJ-1 have fewer spectral bands and the accuracy of acquiring the oil spill coverage only by spectral information is low. In this paper, oil spill occurred in Gulf of Mexico was selected as the research object. The texture characteristics which affect oil spill identification were extracted by Gray Level Co-occurrence Matrix (GLCM) and a decision tree method combined spectral characteristics with texture characteristics was established to extract the oil spill. And comparative analysis with the extraction result of Maximum Likelihood Classification (MLC) was performed. Results showed that, compared with MLC, the decision tree classification raised the user accuracy and the producer accuracy of extracting oil spill by 11.85% and 4.28%, respectively.

KW - Decision tree

KW - Gulf of Mexico

KW - HJ-1

KW - Oil spill

KW - Texture characteristic

UR - http://www.scopus.com/inward/record.url?scp=84907405713&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84907405713&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84907405713

SN - 9789078677772

SP - 756

EP - 759

BT - International Conference on Remote Sensing, Environment and Transportation Engineering, RSETE 2013

PB - Atlantis Press

ER -