Convolutional neural networks for rice yield estimation using MODIS and weather data: A case study for South Korea

Jong Won Ma, Cong Hieu Nguyen, Kyungdo Lee, Joon Heo

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

In South Korea, paddy rice has been consumed over the entire region and it is the main source of income for farmers, thus mathematical model for the estimation of rice yield is required for such decision-making processes in agriculture. The objectives of our study are to: (1) develop rice yield estimation model using Convolutional Neural Networks(CNN), (2) choose hyper-parameters for the model which show the best performance and (3) investigate whether CNN model can effectively predict the rice yield by the comparison with the model using Artificial Neural Networks(ANN). Weather and MODIS(The MOderate Resolution Imaging Spectroradiometer) products from April to September in year 2000∼2013 were used for the rice yield estimation models and crossvalidation was implemented for the accuracy assessment. The CNN and ANN models showed Root Mean Square Error(RMSE) of 36.10kg/10a, 48.61kg/10a based on rice points, respectively and 31.30kg/10a, 39.31kg/10a based on 'Si-Gun-Gu' districts, respectively. The CNN models outperformed ANN models and its possibility of application for the field of rice yield estimation in South Korea was proved.

Original languageEnglish
Pages (from-to)525-534
Number of pages10
JournalJournal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
Volume34
Issue number5
DOIs
Publication statusPublished - 2016 Oct 1

Fingerprint

MODIS
rice
weather
artificial neural network
accuracy assessment
decision making
income
agriculture

All Science Journal Classification (ASJC) codes

  • Earth and Planetary Sciences(all)

Cite this

@article{b66ed525d3404a3788d9795bfbc3a4d0,
title = "Convolutional neural networks for rice yield estimation using MODIS and weather data: A case study for South Korea",
abstract = "In South Korea, paddy rice has been consumed over the entire region and it is the main source of income for farmers, thus mathematical model for the estimation of rice yield is required for such decision-making processes in agriculture. The objectives of our study are to: (1) develop rice yield estimation model using Convolutional Neural Networks(CNN), (2) choose hyper-parameters for the model which show the best performance and (3) investigate whether CNN model can effectively predict the rice yield by the comparison with the model using Artificial Neural Networks(ANN). Weather and MODIS(The MOderate Resolution Imaging Spectroradiometer) products from April to September in year 2000∼2013 were used for the rice yield estimation models and crossvalidation was implemented for the accuracy assessment. The CNN and ANN models showed Root Mean Square Error(RMSE) of 36.10kg/10a, 48.61kg/10a based on rice points, respectively and 31.30kg/10a, 39.31kg/10a based on 'Si-Gun-Gu' districts, respectively. The CNN models outperformed ANN models and its possibility of application for the field of rice yield estimation in South Korea was proved.",
author = "Ma, {Jong Won} and Nguyen, {Cong Hieu} and Kyungdo Lee and Joon Heo",
year = "2016",
month = "10",
day = "1",
doi = "10.7848/ksgpc.2016.34.5.525",
language = "English",
volume = "34",
pages = "525--534",
journal = "Journal of the Korean Society of Surveying Geodesy Photogrammetry and Cartography",
issn = "1598-4850",
publisher = "Korean Society of Surveying, Geodesy, Photogrammetry and Cartography",
number = "5",

}

Convolutional neural networks for rice yield estimation using MODIS and weather data : A case study for South Korea. / Ma, Jong Won; Nguyen, Cong Hieu; Lee, Kyungdo; Heo, Joon.

In: Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography, Vol. 34, No. 5, 01.10.2016, p. 525-534.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Convolutional neural networks for rice yield estimation using MODIS and weather data

T2 - A case study for South Korea

AU - Ma, Jong Won

AU - Nguyen, Cong Hieu

AU - Lee, Kyungdo

AU - Heo, Joon

PY - 2016/10/1

Y1 - 2016/10/1

N2 - In South Korea, paddy rice has been consumed over the entire region and it is the main source of income for farmers, thus mathematical model for the estimation of rice yield is required for such decision-making processes in agriculture. The objectives of our study are to: (1) develop rice yield estimation model using Convolutional Neural Networks(CNN), (2) choose hyper-parameters for the model which show the best performance and (3) investigate whether CNN model can effectively predict the rice yield by the comparison with the model using Artificial Neural Networks(ANN). Weather and MODIS(The MOderate Resolution Imaging Spectroradiometer) products from April to September in year 2000∼2013 were used for the rice yield estimation models and crossvalidation was implemented for the accuracy assessment. The CNN and ANN models showed Root Mean Square Error(RMSE) of 36.10kg/10a, 48.61kg/10a based on rice points, respectively and 31.30kg/10a, 39.31kg/10a based on 'Si-Gun-Gu' districts, respectively. The CNN models outperformed ANN models and its possibility of application for the field of rice yield estimation in South Korea was proved.

AB - In South Korea, paddy rice has been consumed over the entire region and it is the main source of income for farmers, thus mathematical model for the estimation of rice yield is required for such decision-making processes in agriculture. The objectives of our study are to: (1) develop rice yield estimation model using Convolutional Neural Networks(CNN), (2) choose hyper-parameters for the model which show the best performance and (3) investigate whether CNN model can effectively predict the rice yield by the comparison with the model using Artificial Neural Networks(ANN). Weather and MODIS(The MOderate Resolution Imaging Spectroradiometer) products from April to September in year 2000∼2013 were used for the rice yield estimation models and crossvalidation was implemented for the accuracy assessment. The CNN and ANN models showed Root Mean Square Error(RMSE) of 36.10kg/10a, 48.61kg/10a based on rice points, respectively and 31.30kg/10a, 39.31kg/10a based on 'Si-Gun-Gu' districts, respectively. The CNN models outperformed ANN models and its possibility of application for the field of rice yield estimation in South Korea was proved.

UR - http://www.scopus.com/inward/record.url?scp=85017531248&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85017531248&partnerID=8YFLogxK

U2 - 10.7848/ksgpc.2016.34.5.525

DO - 10.7848/ksgpc.2016.34.5.525

M3 - Article

AN - SCOPUS:85017531248

VL - 34

SP - 525

EP - 534

JO - Journal of the Korean Society of Surveying Geodesy Photogrammetry and Cartography

JF - Journal of the Korean Society of Surveying Geodesy Photogrammetry and Cartography

SN - 1598-4850

IS - 5

ER -