A support vector machine to identify irrigated crop types using time-series Landsat NDVI data

Baojuan Zheng, Soe Myint, Prasad S. Thenkabail, Rimjhim Aggarwal

Research output: Contribution to journalArticle

111 Citations (Scopus)

Abstract

Site-specific information of crop types is required for many agro-environmental assessments. The study investigated the potential of support vector machines (SVMs) in discriminating various crop types in a complex cropping system in the Phoenix Active Management Area. We applied SVMs to Landsat time-series Normalized Difference Vegetation Index (NDVI) data using training datasets selected by two different approaches: Stratified random approach and intelligent selection approach using local knowledge. The SVM models effectively classified nine major crop types with overall accuracies of >86% for both training datasets. Our results showed that the intelligent selection approach was able to reduce the training set size and achieved higher overall classification accuracy than the stratified random approach. The intelligent selection approach is particularly useful when the availability of reference data is limited and unbalanced among different classes. The study demonstrated the potential of utilizing multi-temporal Landsat imagery to systematically monitor crop types and cropping patterns over time in arid and semi-arid regions.

Original languageEnglish (US)
Pages (from-to)103-112
Number of pages10
JournalInternational Journal of Applied Earth Observation and Geoinformation
Volume34
Issue number1
DOIs
StatePublished - 2015

Fingerprint

NDVI
Crops
Landsat
Support vector machines
Time series
time series
crop
Arid regions
cropping practice
Large scale systems
traditional knowledge
Availability
environmental assessment
semiarid region
support vector machine
imagery

Keywords

  • Crop classification
  • Landsat
  • NDVI
  • Support vector machines
  • SVM

ASJC Scopus subject areas

  • Computers in Earth Sciences
  • Earth-Surface Processes
  • Global and Planetary Change
  • Management, Monitoring, Policy and Law

Cite this

A support vector machine to identify irrigated crop types using time-series Landsat NDVI data. / Zheng, Baojuan; Myint, Soe; Thenkabail, Prasad S.; Aggarwal, Rimjhim.

In: International Journal of Applied Earth Observation and Geoinformation, Vol. 34, No. 1, 2015, p. 103-112.

Research output: Contribution to journalArticle

@article{61c7d35b5d0543a2a5f394b7c92a613f,
title = "A support vector machine to identify irrigated crop types using time-series Landsat NDVI data",
abstract = "Site-specific information of crop types is required for many agro-environmental assessments. The study investigated the potential of support vector machines (SVMs) in discriminating various crop types in a complex cropping system in the Phoenix Active Management Area. We applied SVMs to Landsat time-series Normalized Difference Vegetation Index (NDVI) data using training datasets selected by two different approaches: Stratified random approach and intelligent selection approach using local knowledge. The SVM models effectively classified nine major crop types with overall accuracies of >86{\%} for both training datasets. Our results showed that the intelligent selection approach was able to reduce the training set size and achieved higher overall classification accuracy than the stratified random approach. The intelligent selection approach is particularly useful when the availability of reference data is limited and unbalanced among different classes. The study demonstrated the potential of utilizing multi-temporal Landsat imagery to systematically monitor crop types and cropping patterns over time in arid and semi-arid regions.",
keywords = "Crop classification, Landsat, NDVI, Support vector machines, SVM",
author = "Baojuan Zheng and Soe Myint and Thenkabail, {Prasad S.} and Rimjhim Aggarwal",
year = "2015",
doi = "10.1016/j.jag.2014.07.002",
language = "English (US)",
volume = "34",
pages = "103--112",
journal = "International Journal of Applied Earth Observation and Geoinformation",
issn = "1569-8432",
publisher = "Elsevier",
number = "1",

}

TY - JOUR

T1 - A support vector machine to identify irrigated crop types using time-series Landsat NDVI data

AU - Zheng, Baojuan

AU - Myint, Soe

AU - Thenkabail, Prasad S.

AU - Aggarwal, Rimjhim

PY - 2015

Y1 - 2015

N2 - Site-specific information of crop types is required for many agro-environmental assessments. The study investigated the potential of support vector machines (SVMs) in discriminating various crop types in a complex cropping system in the Phoenix Active Management Area. We applied SVMs to Landsat time-series Normalized Difference Vegetation Index (NDVI) data using training datasets selected by two different approaches: Stratified random approach and intelligent selection approach using local knowledge. The SVM models effectively classified nine major crop types with overall accuracies of >86% for both training datasets. Our results showed that the intelligent selection approach was able to reduce the training set size and achieved higher overall classification accuracy than the stratified random approach. The intelligent selection approach is particularly useful when the availability of reference data is limited and unbalanced among different classes. The study demonstrated the potential of utilizing multi-temporal Landsat imagery to systematically monitor crop types and cropping patterns over time in arid and semi-arid regions.

AB - Site-specific information of crop types is required for many agro-environmental assessments. The study investigated the potential of support vector machines (SVMs) in discriminating various crop types in a complex cropping system in the Phoenix Active Management Area. We applied SVMs to Landsat time-series Normalized Difference Vegetation Index (NDVI) data using training datasets selected by two different approaches: Stratified random approach and intelligent selection approach using local knowledge. The SVM models effectively classified nine major crop types with overall accuracies of >86% for both training datasets. Our results showed that the intelligent selection approach was able to reduce the training set size and achieved higher overall classification accuracy than the stratified random approach. The intelligent selection approach is particularly useful when the availability of reference data is limited and unbalanced among different classes. The study demonstrated the potential of utilizing multi-temporal Landsat imagery to systematically monitor crop types and cropping patterns over time in arid and semi-arid regions.

KW - Crop classification

KW - Landsat

KW - NDVI

KW - Support vector machines

KW - SVM

UR - http://www.scopus.com/inward/record.url?scp=84908440807&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84908440807&partnerID=8YFLogxK

U2 - 10.1016/j.jag.2014.07.002

DO - 10.1016/j.jag.2014.07.002

M3 - Article

AN - SCOPUS:84908440807

VL - 34

SP - 103

EP - 112

JO - International Journal of Applied Earth Observation and Geoinformation

JF - International Journal of Applied Earth Observation and Geoinformation

SN - 1569-8432

IS - 1

ER -