Classification of mangroves vegetation species using texture analysis on RapidEye satellite imagery

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

4 Citations (Scopus)

Abstract

Mangroves are unique ecosystem structures that are typically made up of salt tolerant species of vegetation that can be found in tropical and subtropical climate country. Mangrove ecosystem plays important role and also is known as highly productive ecosystem with high diversity of flora and fauna. However, these ecosystems have been declining over time due to the various kinds of direct and indirect pressures. Thus, there is an increasing need to monitor and assess this ecosystem for better conservation and management efforts. The multispectral RapidEye satellite image was used to identify the mangrove vegetation species within the Matang Mangrove Forest Reserve in Perak, Malaysia using texture analysis. Classification was implemented using the maximum likelihood classifier (MLC) method. Total of eleven main mangrove species were found in the satellite image of the study site which includes Rhizophora mucronata, Rhizophora apiculata, Bruguiera parviflora, Bruguiera cylindrica, Bruguiera gymnorrhiza, Avicennia alba, Avicennia officinalis, Sonneratia alba, Sonneratia caseolaris, Sonneratia ovata and Xylocarpus granatum. The classification results showed that the textured image produced high overall classification assessment recorded at 84% and kappa statistic of 0.8016. Meanwhile, the non-textured image produces 80% of overall accuracy and kappa statistic of 0.7061. The classification result indicated the capability of high resolution satellite image to classify the mangrove species and inclusion of texture information in the classification increased the classification accuracy.

Original languageEnglish
Title of host publicationAIP Conference Proceedings
Pages480-486
Number of pages7
Volume1571
DOIs
Publication statusPublished - 2013
Event2013 UKM Faculty of Science and Technology Post-Graduate Colloquium - Selangor
Duration: 3 Jul 20134 Jul 2013

Other

Other2013 UKM Faculty of Science and Technology Post-Graduate Colloquium
CitySelangor
Period3/7/134/7/13

Fingerprint

satellite imagery
vegetation
ecosystems
textures
statistics
plants (botany)
Malaysia
classifiers
climate
animals
conservation
inclusions
salts
high resolution

Keywords

  • Mangrove vegetation species
  • Matang mangrove forest reserve
  • MLC
  • RapidEye satellite image

ASJC Scopus subject areas

  • Physics and Astronomy(all)

Cite this

Classification of mangroves vegetation species using texture analysis on RapidEye satellite imagery. / Roslani, M. A.; Ahmad Mustapha, Muzzneena; Lihan, Tukimat; Wan Ahmad, Wan Juliana.

AIP Conference Proceedings. Vol. 1571 2013. p. 480-486.

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

Roslani, MA, Ahmad Mustapha, M, Lihan, T & Wan Ahmad, WJ 2013, Classification of mangroves vegetation species using texture analysis on RapidEye satellite imagery. in AIP Conference Proceedings. vol. 1571, pp. 480-486, 2013 UKM Faculty of Science and Technology Post-Graduate Colloquium, Selangor, 3/7/13. https://doi.org/10.1063/1.4858701
@inproceedings{39204f15af12419cb9d965080d3a2a1f,
title = "Classification of mangroves vegetation species using texture analysis on RapidEye satellite imagery",
abstract = "Mangroves are unique ecosystem structures that are typically made up of salt tolerant species of vegetation that can be found in tropical and subtropical climate country. Mangrove ecosystem plays important role and also is known as highly productive ecosystem with high diversity of flora and fauna. However, these ecosystems have been declining over time due to the various kinds of direct and indirect pressures. Thus, there is an increasing need to monitor and assess this ecosystem for better conservation and management efforts. The multispectral RapidEye satellite image was used to identify the mangrove vegetation species within the Matang Mangrove Forest Reserve in Perak, Malaysia using texture analysis. Classification was implemented using the maximum likelihood classifier (MLC) method. Total of eleven main mangrove species were found in the satellite image of the study site which includes Rhizophora mucronata, Rhizophora apiculata, Bruguiera parviflora, Bruguiera cylindrica, Bruguiera gymnorrhiza, Avicennia alba, Avicennia officinalis, Sonneratia alba, Sonneratia caseolaris, Sonneratia ovata and Xylocarpus granatum. The classification results showed that the textured image produced high overall classification assessment recorded at 84{\%} and kappa statistic of 0.8016. Meanwhile, the non-textured image produces 80{\%} of overall accuracy and kappa statistic of 0.7061. The classification result indicated the capability of high resolution satellite image to classify the mangrove species and inclusion of texture information in the classification increased the classification accuracy.",
keywords = "Mangrove vegetation species, Matang mangrove forest reserve, MLC, RapidEye satellite image",
author = "Roslani, {M. A.} and {Ahmad Mustapha}, Muzzneena and Tukimat Lihan and {Wan Ahmad}, {Wan Juliana}",
year = "2013",
doi = "10.1063/1.4858701",
language = "English",
isbn = "9780735411999",
volume = "1571",
pages = "480--486",
booktitle = "AIP Conference Proceedings",

}

TY - GEN

T1 - Classification of mangroves vegetation species using texture analysis on RapidEye satellite imagery

AU - Roslani, M. A.

AU - Ahmad Mustapha, Muzzneena

AU - Lihan, Tukimat

AU - Wan Ahmad, Wan Juliana

PY - 2013

Y1 - 2013

N2 - Mangroves are unique ecosystem structures that are typically made up of salt tolerant species of vegetation that can be found in tropical and subtropical climate country. Mangrove ecosystem plays important role and also is known as highly productive ecosystem with high diversity of flora and fauna. However, these ecosystems have been declining over time due to the various kinds of direct and indirect pressures. Thus, there is an increasing need to monitor and assess this ecosystem for better conservation and management efforts. The multispectral RapidEye satellite image was used to identify the mangrove vegetation species within the Matang Mangrove Forest Reserve in Perak, Malaysia using texture analysis. Classification was implemented using the maximum likelihood classifier (MLC) method. Total of eleven main mangrove species were found in the satellite image of the study site which includes Rhizophora mucronata, Rhizophora apiculata, Bruguiera parviflora, Bruguiera cylindrica, Bruguiera gymnorrhiza, Avicennia alba, Avicennia officinalis, Sonneratia alba, Sonneratia caseolaris, Sonneratia ovata and Xylocarpus granatum. The classification results showed that the textured image produced high overall classification assessment recorded at 84% and kappa statistic of 0.8016. Meanwhile, the non-textured image produces 80% of overall accuracy and kappa statistic of 0.7061. The classification result indicated the capability of high resolution satellite image to classify the mangrove species and inclusion of texture information in the classification increased the classification accuracy.

AB - Mangroves are unique ecosystem structures that are typically made up of salt tolerant species of vegetation that can be found in tropical and subtropical climate country. Mangrove ecosystem plays important role and also is known as highly productive ecosystem with high diversity of flora and fauna. However, these ecosystems have been declining over time due to the various kinds of direct and indirect pressures. Thus, there is an increasing need to monitor and assess this ecosystem for better conservation and management efforts. The multispectral RapidEye satellite image was used to identify the mangrove vegetation species within the Matang Mangrove Forest Reserve in Perak, Malaysia using texture analysis. Classification was implemented using the maximum likelihood classifier (MLC) method. Total of eleven main mangrove species were found in the satellite image of the study site which includes Rhizophora mucronata, Rhizophora apiculata, Bruguiera parviflora, Bruguiera cylindrica, Bruguiera gymnorrhiza, Avicennia alba, Avicennia officinalis, Sonneratia alba, Sonneratia caseolaris, Sonneratia ovata and Xylocarpus granatum. The classification results showed that the textured image produced high overall classification assessment recorded at 84% and kappa statistic of 0.8016. Meanwhile, the non-textured image produces 80% of overall accuracy and kappa statistic of 0.7061. The classification result indicated the capability of high resolution satellite image to classify the mangrove species and inclusion of texture information in the classification increased the classification accuracy.

KW - Mangrove vegetation species

KW - Matang mangrove forest reserve

KW - MLC

KW - RapidEye satellite image

UR - http://www.scopus.com/inward/record.url?scp=84897750419&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897750419&partnerID=8YFLogxK

U2 - 10.1063/1.4858701

DO - 10.1063/1.4858701

M3 - Conference contribution

AN - SCOPUS:84897750419

SN - 9780735411999

VL - 1571

SP - 480

EP - 486

BT - AIP Conference Proceedings

ER -