Current Science

Open Access Open Access  Restricted Access Subscription Access

Crop production estimation using deep learning technique


Affiliations
1 CSIR-Fourth Paradigm Institute, Bengaluru 560 037, India and Academy of Scientific and Innovative Research, Ghaziabad 201 002, India, India
 

Reliable estimation of crop requirement and production in advance, help policy makers to adopt timely decision for trade as export–import, which is a basic building block to assure food security of a country. A powerful and robust algorithm is essential to predict the future demand and production of a particular crop for subsequent years. Deep learning methods are used successfully in solving different prediction pro­blems of various applications. This study attempts to design an efficient AI based technique specifically using long short-term memory, a deep learning approach for estimation of crop production using crop production information of neighbouring countries, which are part of the South Asian monsoon system. Detailed sensitivity analysis is conducted to identify the optimal combination of crop production of neighbouring countries that directly and indirectly impact the crop production of India. Here, we designed and developed a predictive model for rice production of India with lead time of one year using deep learning technique. Along with that, as there are significant influences of local climate (i.e. rainfall data) on crop production, that information was also considered along with crop production of neighbouring countries. The results indi­cated that local and regional scale parameters jointly improve the prediction capability for future years. Capability of the proposed model was validated with export–import data on crop of India and neighbouring countries, and the validation result showed that our proposed technique was efficient and robust in nature

Keywords

Artificial intelligence, crop production model, deep neural networks, long short-term memory, sensitivity analysis.
User
Notifications
Font Size

  • Horie, T., Yajima, M. and Nakagawa, H., Yield forecasting. Agric. Syst., 1992, 40, 211–236; doi:10.1016/0308-521X(92)90022-G.

  • Carfagna, E. and Gallego, F. J., Using remote sensing for agricultural statistics. Int. Stat. Rev., 2005, 73(3), 389–404.

  • Awad, M. M., Toward precision in crop yield estimation using remote sensing and optimization techniques. Agriculture, 2019, 9(3), 54; https://doi.org/10.3390/agriculture9030054.

  • Paliwal, A. and Jain, M., The accuracy of self-reported crop yield estimates and their ability to train remote sensing algorithms. Front. Sustain. Food Syst., 2020; doi:10.3389/fsufs.2020.00025.

  • You, J., Li, X., Low, M., Lobell, D. and Ermon, S., Deep Gaussian process for crop yield prediction based on remote sensing data. In Thirty-First AAAI Conference on Artificial Intelligence, San Francisco, USA, 2017, pp. 4559–4566.

  • Russello, H., Convolutional neural networks for crop yield prediction using satellite images. IBM Center for Advanced Studies, Berelux, 2018.

  • http://www.fao.org/about/en

  • Sarker, I. H., Machine learning: algorithms, real-world applications and research directions. SN Comput. Sci., 2021, 2, 160; https:// doi.org/10.1007/s42979-021-00592-x.

  • Hochreiter, S. and Schmidhuber, J., Long short-term memory. Neural Comput., 1997, 9(8), 1735–1780; doi:10.1162/neco.1997.9.8.1735.

  • Solow, A. et al., The value of improved ENSO prediction to US agriculture. Climatic Change, 1998, 39, 47–60.

  • Jones, J. W., Hansem, J. W., Royce, F. S. and Messina, C. D., Potential benefits of climate forecasting in agriculture. Agric. Ecosyst. Environ., 2000, 82, 169–184.

  • Hansen, J. W., Applying seasonal climate prediction to agriculture production. Agric. Syst., 2002, 74(3), 305–307.

  • Prasanna, V., Impact of monsoon rainfall on the total food grain yield over India. Proc. Indian Acad. Sci. (Earth Planet. Sci.), 2014, 112, 529–558.

  • Rahman, M. A. et al., Impacts of temperature and rainfall variation on rice productivity in major ecosystems of Bangladesh. Agric Food Secur., 2017, 6, 10; https://doi.org/10.1186/s40066-017-0089-5.

  • Baigorria, G. A., Jones, J. W., Shin, D. W., Mishra, A. and O’Brien, J. J., Assessing uncertainties in crop model simulations using daily bias-corrected regional circulation model outputs. Clim. Res., 2007, 34, 211–222.

  • Cantelaube, P. and Terres, J. M., Seasonal weather forecasts for crop yield modelling in Europe. Tellus A: Dyn. Meteorol. Oceanogr., 2005, 57A, 476–487.


Abstract Views: 403

PDF Views: 140




  • Crop production estimation using deep learning technique

Abstract Views: 403  |  PDF Views: 140

Authors

Ashapurna Marndi
CSIR-Fourth Paradigm Institute, Bengaluru 560 037, India and Academy of Scientific and Innovative Research, Ghaziabad 201 002, India, India
K. V. Ramesh
CSIR-Fourth Paradigm Institute, Bengaluru 560 037, India and Academy of Scientific and Innovative Research, Ghaziabad 201 002, India, India
G. K. Patra
CSIR-Fourth Paradigm Institute, Bengaluru 560 037, India and Academy of Scientific and Innovative Research, Ghaziabad 201 002, India, India

Abstract


Reliable estimation of crop requirement and production in advance, help policy makers to adopt timely decision for trade as export–import, which is a basic building block to assure food security of a country. A powerful and robust algorithm is essential to predict the future demand and production of a particular crop for subsequent years. Deep learning methods are used successfully in solving different prediction pro­blems of various applications. This study attempts to design an efficient AI based technique specifically using long short-term memory, a deep learning approach for estimation of crop production using crop production information of neighbouring countries, which are part of the South Asian monsoon system. Detailed sensitivity analysis is conducted to identify the optimal combination of crop production of neighbouring countries that directly and indirectly impact the crop production of India. Here, we designed and developed a predictive model for rice production of India with lead time of one year using deep learning technique. Along with that, as there are significant influences of local climate (i.e. rainfall data) on crop production, that information was also considered along with crop production of neighbouring countries. The results indi­cated that local and regional scale parameters jointly improve the prediction capability for future years. Capability of the proposed model was validated with export–import data on crop of India and neighbouring countries, and the validation result showed that our proposed technique was efficient and robust in nature

Keywords


Artificial intelligence, crop production model, deep neural networks, long short-term memory, sensitivity analysis.

References





DOI: https://doi.org/10.18520/cs%2Fv121%2Fi8%2F1073-1079