Aarhus University Seal / Aarhus Universitets segl

Mogens Humlekrog Greve

A Multievidence Approach for Crop Discrimination Using Multitemporal WorldView-2 Imagery

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avisTidsskriftartikelForskningpeer review

Standard

A Multievidence Approach for Crop Discrimination Using Multitemporal WorldView-2 Imagery. / Chellasamy, Menaka; Zielinski, Rafal Tomasz; Greve, Mogens Humlekrog.

I: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (J-STARS), Bind 7, Nr. 8, 2349945, 2014.

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avisTidsskriftartikelForskningpeer review

Harvard

Chellasamy, M, Zielinski, RT & Greve, MH 2014, 'A Multievidence Approach for Crop Discrimination Using Multitemporal WorldView-2 Imagery', IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (J-STARS), bind 7, nr. 8, 2349945. https://doi.org/10.1109/JSTARS.2014.2349945

APA

Chellasamy, M., Zielinski, R. T., & Greve, M. H. (2014). A Multievidence Approach for Crop Discrimination Using Multitemporal WorldView-2 Imagery. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (J-STARS), 7(8), [2349945]. https://doi.org/10.1109/JSTARS.2014.2349945

CBE

Chellasamy M, Zielinski RT, Greve MH. 2014. A Multievidence Approach for Crop Discrimination Using Multitemporal WorldView-2 Imagery. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (J-STARS). 7(8):Article 2349945. https://doi.org/10.1109/JSTARS.2014.2349945

MLA

Chellasamy, Menaka, Rafal Tomasz Zielinski, og Mogens Humlekrog Greve. "A Multievidence Approach for Crop Discrimination Using Multitemporal WorldView-2 Imagery". IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (J-STARS). 2014. 7(8). https://doi.org/10.1109/JSTARS.2014.2349945

Vancouver

Chellasamy M, Zielinski RT, Greve MH. A Multievidence Approach for Crop Discrimination Using Multitemporal WorldView-2 Imagery. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (J-STARS). 2014;7(8). 2349945. https://doi.org/10.1109/JSTARS.2014.2349945

Author

Chellasamy, Menaka ; Zielinski, Rafal Tomasz ; Greve, Mogens Humlekrog. / A Multievidence Approach for Crop Discrimination Using Multitemporal WorldView-2 Imagery. I: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (J-STARS). 2014 ; Bind 7, Nr. 8.

Bibtex

@article{050fc52024224b099a5ed58f02d461e7,
title = "A Multievidence Approach for Crop Discrimination Using Multitemporal WorldView-2 Imagery",
abstract = "Despite using multiple input datasets for effective crop classification, it is important to select an appropriate method that efficiently integrates these multiple datasets to produce accurate classification results. In this paper, we present an endorsement theory-based crop classification approach that considers the qualitative information, in terms of prediction probabilities, from different input datasets and integrates them efficiently to produce final classification results. Three different input datasets are used in this study: 1) spectral; 2) texture; and 3) indices from multitemporal (spring, early summer) WorldView-2 multispectral imagery. A multilayer perceptron classifier is trained with the multitemporal datasets separately using a backpropagation learning algorithm, and prediction probabilities are produced for each pixel as evidence against each crop class. An integration rule based on endorsement theory is applied to these multiple evidence by considering their individual contribution, and the most probable class of a pixel is identified. Integration of the three multidate datasets by the proposed method is found to produce higher overall classification accuracy (91.2%) when compared to conventional winner-takes-all approach (89%). In order to determine which individual dataset is more useful for crop discrimination, the dataset{\textquoteright}s performance is compared using evidence and contributions produced in the proposed integration method for four selected crops, for both single- and multidate. The results of this analyses showed that seasonal textures information outperformed both spectral and indices. To verify this finding, results of individual dataset classification are examined. The highest overall classification accuracy of 88.8% is achieved by the use of multidate texture, where multidate spectral and indices resulted in 86.3% and 84.4%, respectively.",
keywords = "crop discrimination, WorldView-2 (WV2), enforsement theory (ET), multitemporal datasets, neural classifier",
author = "Menaka Chellasamy and Zielinski, {Rafal Tomasz} and Greve, {Mogens Humlekrog}",
year = "2014",
doi = "10.1109/JSTARS.2014.2349945",
language = "English",
volume = "7",
journal = "IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (J-STARS)",
issn = "1939-1404",
number = "8",

}

RIS

TY - JOUR

T1 - A Multievidence Approach for Crop Discrimination Using Multitemporal WorldView-2 Imagery

AU - Chellasamy, Menaka

AU - Zielinski, Rafal Tomasz

AU - Greve, Mogens Humlekrog

PY - 2014

Y1 - 2014

N2 - Despite using multiple input datasets for effective crop classification, it is important to select an appropriate method that efficiently integrates these multiple datasets to produce accurate classification results. In this paper, we present an endorsement theory-based crop classification approach that considers the qualitative information, in terms of prediction probabilities, from different input datasets and integrates them efficiently to produce final classification results. Three different input datasets are used in this study: 1) spectral; 2) texture; and 3) indices from multitemporal (spring, early summer) WorldView-2 multispectral imagery. A multilayer perceptron classifier is trained with the multitemporal datasets separately using a backpropagation learning algorithm, and prediction probabilities are produced for each pixel as evidence against each crop class. An integration rule based on endorsement theory is applied to these multiple evidence by considering their individual contribution, and the most probable class of a pixel is identified. Integration of the three multidate datasets by the proposed method is found to produce higher overall classification accuracy (91.2%) when compared to conventional winner-takes-all approach (89%). In order to determine which individual dataset is more useful for crop discrimination, the dataset’s performance is compared using evidence and contributions produced in the proposed integration method for four selected crops, for both single- and multidate. The results of this analyses showed that seasonal textures information outperformed both spectral and indices. To verify this finding, results of individual dataset classification are examined. The highest overall classification accuracy of 88.8% is achieved by the use of multidate texture, where multidate spectral and indices resulted in 86.3% and 84.4%, respectively.

AB - Despite using multiple input datasets for effective crop classification, it is important to select an appropriate method that efficiently integrates these multiple datasets to produce accurate classification results. In this paper, we present an endorsement theory-based crop classification approach that considers the qualitative information, in terms of prediction probabilities, from different input datasets and integrates them efficiently to produce final classification results. Three different input datasets are used in this study: 1) spectral; 2) texture; and 3) indices from multitemporal (spring, early summer) WorldView-2 multispectral imagery. A multilayer perceptron classifier is trained with the multitemporal datasets separately using a backpropagation learning algorithm, and prediction probabilities are produced for each pixel as evidence against each crop class. An integration rule based on endorsement theory is applied to these multiple evidence by considering their individual contribution, and the most probable class of a pixel is identified. Integration of the three multidate datasets by the proposed method is found to produce higher overall classification accuracy (91.2%) when compared to conventional winner-takes-all approach (89%). In order to determine which individual dataset is more useful for crop discrimination, the dataset’s performance is compared using evidence and contributions produced in the proposed integration method for four selected crops, for both single- and multidate. The results of this analyses showed that seasonal textures information outperformed both spectral and indices. To verify this finding, results of individual dataset classification are examined. The highest overall classification accuracy of 88.8% is achieved by the use of multidate texture, where multidate spectral and indices resulted in 86.3% and 84.4%, respectively.

KW - crop discrimination

KW - WorldView-2 (WV2)

KW - enforsement theory (ET)

KW - multitemporal datasets

KW - neural classifier

U2 - 10.1109/JSTARS.2014.2349945

DO - 10.1109/JSTARS.2014.2349945

M3 - Journal article

VL - 7

JO - IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (J-STARS)

JF - IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (J-STARS)

SN - 1939-1404

IS - 8

M1 - 2349945

ER -