A Caltech Library Service

Estimation of Nitrogen in Rice Crops from UAV-Captured Images

Colorado, Julian D. and Cera-Bornacelli, Natalia and Caldas, Juan S. and Petro, Eliel and Rebolledo, Maria C. and Cuellar, David and Calderon, Francisco and Mondragon, Ivan F. and Jaramillo-Botero, Andres (2020) Estimation of Nitrogen in Rice Crops from UAV-Captured Images. Remote Sensing, 12 (20). Art. No. 3396. ISSN 2072-4292. doi:10.3390/rs12203396.

PDF - Published Version
Creative Commons Attribution.

[img] Archive (ZIP) - Supplemental Material
Creative Commons Attribution.


Use this Persistent URL to link to this item:


Leaf nitrogen (N) directly correlates to chlorophyll production, affecting crop growth and yield. Farmers use soil plant analysis development (SPAD) devices to calculate the amount of chlorophyll present in plants. However, monitoring large-scale crops using SPAD is prohibitively time-consuming and demanding. This paper presents an unmanned aerial vehicle (UAV) solution for estimating leaf N content in rice crops, from multispectral imagery. Our contribution is twofold: (i) a novel trajectory control strategy to reduce the angular wind-induced perturbations that affect image sampling accuracy during UAV flight, and (ii) machine learning models to estimate the canopy N via vegetation indices (VIs) obtained from the aerial imagery. This approach integrates an image processing algorithm using the GrabCut segmentation method with a guided filtering refinement process, to calculate the VIs according to the plots of interest. Three machine learning methods based on multivariable linear regressions (MLR), support vector machines (SVM), and neural networks (NN), were applied and compared through the entire phonological cycle of the crop: vegetative (V), reproductive (R), and ripening (Ri). Correlations were obtained by comparing our methods against an assembled ground-truth of SPAD measurements. The higher N correlations were achieved with NN: 0.98 (V), 0.94 (R), and 0.89 (Ri). We claim that the proposed UAV stabilization control algorithm significantly improves on the N-to-SPAD correlations by minimizing wind perturbations in real-time and reducing the need for offline image corrections.

Item Type:Article
Related URLs:
URLURL TypeDescription File 1 ItemFILE S1: RAW data supporting image segmentation metrics, multispectral imagery used for Machine Learning testing, and nitrogen estimation results available at the Open Science Framework ItemFILE S2: Experimental protocol for crop monitoring
Colorado, Julian D.0000-0002-6925-0126
Rebolledo, Maria C.0000-0002-4369-5688
Calderon, Francisco0000-0001-8681-415X
Mondragon, Ivan F.0000-0002-7828-6681
Jaramillo-Botero, Andres0000-0003-2844-0756
Additional Information:© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Received: 2 July 2020; Accepted: 22 September 2020; Published: 16 October 2020. (This article belongs to the Special Issue UAVs for Vegetation Monitoring) The authors would like to thank all CIAT staff that supported the experiments over the crops located at CIAT headquarters in Palmira, Valle del Cauca, Colombia; in particular to Yolima Ospina and Cecile Grenier for their support in upland and lowland trials. Also, to Carlos Devia from Javeriana University for the insights regarding the structuring of the datasets. This work was funded by the OMICAS program: Optimización Multiescala In-silico de Cultivos Agrícolas Sostenibles (Infraestructura y validación en Arroz y Caña de Azúcar), anchored at the Pontificia Universidad Javeriana in Cali and funded within the Colombian Scientific Ecosystem by The World Bank, the Colombian Ministry of Science, Technology and Innovation, the Colombian Ministry of Education and the Colombian Ministry of Industry and Tourism, and ICETEX under GRANT ID: FP44842-217-2018. Author Contributions: Conceptualization, J.D.C. (Julian D. Colorado), N.C.-B., F.C., and J.S.C. (Juan S. Caldas); methodology, J.D.C. (Julian D. Colorado), M.C.R., F.C., and A.J.-B.; software, N.C.-B., J.S.C. (Juan S. Caldas), and D.C.; validation, J.D.C. (Julian D. Colorado), I.F.M., E.P., D.C., N.C.-B., and J.S.C. (Juan S. Caldas); formal analysis and investigation, J.D.C. (Julian D. Colorado), M.C.R., E.P., F.C., I.F.M., and A.J.-B.; data curation, E.P.; writing—original draft preparation, J.D.C. (Julian D. Colorado) and F.C.; writing—review and editing, A.J.-B., I.F.M., and M.C.R.; supervision, A.J.-B. and J.D.C. (Julian D. Colorado). All authors have read and agreed to the published version of the manuscript. The authors declare no conflict of interest. Supplementary Materials: The following are available online at, FILE S1: RAW data supporting image segmentation metrics, multispectral imagery imagery used for machine learning testing, and nitrogen estimation results are available at the Open Science Framework: FILE S2: Experimental protocol for crop monitoring available at VIDEO S3: The video is available in the online version of this article. The video accompanying this paper illustrates the steps performed during the experiments.
Funding AgencyGrant Number
Ministry of Science, Technology and Innovation (Colombia)UNSPECIFIED
Ministry of Education (Colombia)UNSPECIFIED
Ministry of Industry and Tourism (Colombia)UNSPECIFIED
Instituto Colombiano de Crédito Educativo y Estudios Técnicos en el Exterior (ICETEX)FP44842-217-2018
Subject Keywords:UAV; machine learning; plant nitrogen estimation; multispectral imagery; vegetation index; image segmentation
Issue or Number:20
Record Number:CaltechAUTHORS:20201016-131847873
Persistent URL:
Official Citation:Colorado, J.; Cera-Bornacelli, N.; Caldas, J.; Petro, E.; Rebolledo, C.; Cuellar, D.; Calderon, F.; Mondragon, I.; Jaramillo-Botero, A. Estimation of Nitrogen in Rice Crops from UAV-Captured Images. Remote Sens. 2020, 12, 3396.
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:106116
Deposited By: George Porter
Deposited On:16 Oct 2020 21:19
Last Modified:16 Nov 2021 18:50

Repository Staff Only: item control page