Estimating and evaluating the rice cluster distribution uniformity with UAV-based images

Abstract The uniformity of the rice cluster distribution in the field affects population quality and the precise management of pesticides and fertilizers. However, there is no appropriate technical system for estimating and evaluating the uniformity at present. For that reason, a method based on unm...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Xiaohui Wang, Qiyuan Tang, Zhaozhong Chen, Youyi Luo, Hongyu Fu, Xumeng Li
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/5325ffc30c314994a05fbfc6bae22aa4
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:5325ffc30c314994a05fbfc6bae22aa4
record_format	dspace
spelling	oai:doaj.org-article:5325ffc30c314994a05fbfc6bae22aa42021-11-08T10:49:20ZEstimating and evaluating the rice cluster distribution uniformity with UAV-based images10.1038/s41598-021-01044-52045-2322https://doaj.org/article/5325ffc30c314994a05fbfc6bae22aa42021-11-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-01044-5https://doaj.org/toc/2045-2322Abstract The uniformity of the rice cluster distribution in the field affects population quality and the precise management of pesticides and fertilizers. However, there is no appropriate technical system for estimating and evaluating the uniformity at present. For that reason, a method based on unmanned aerial vehicle (UAV images) is proposed to estimate and evaluate the uniformity in this present study. This method includes rice cluster recognition and location determination based on the RGB color characteristics of the seedlings of aerial images, region segmentation considering the rice clusters based on Voronoi Diagram, and uniformity index definition for evaluating the rice cluster distribution based on the variation coefficient. The results indicate the rice cluster recognition attains a high precision, with the precision, accuracy, recall, and F1-score of rice cluster recognition reaching > 95%, 97%, 97%, 95%, and 96%, respectively. The rice cluster location error is small and obeys the gamma (3.00, 0.54) distribution (mean error, 1.62 cm). The uniformity index is reasonable for evaluating the rice cluster distribution verified via simulation. As a whole process, the estimating method is sufficiently high accuracy with relative error less than 0.01% over the manual labeling method. Therefore, this method based on UAV images is feasible, convenient, technologically advanced, inexpensive, and highly precision for the estimation and evaluation of the rice cluster distribution uniformity. However, the evaluation application indicates that there is much room for improvement in terms of the uniformity of mechanized paddy field transplanting in South China.Xiaohui WangQiyuan TangZhaozhong ChenYouyi LuoHongyu FuXumeng LiNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-11 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Xiaohui Wang Qiyuan Tang Zhaozhong Chen Youyi Luo Hongyu Fu Xumeng Li Estimating and evaluating the rice cluster distribution uniformity with UAV-based images
description	Abstract The uniformity of the rice cluster distribution in the field affects population quality and the precise management of pesticides and fertilizers. However, there is no appropriate technical system for estimating and evaluating the uniformity at present. For that reason, a method based on unmanned aerial vehicle (UAV images) is proposed to estimate and evaluate the uniformity in this present study. This method includes rice cluster recognition and location determination based on the RGB color characteristics of the seedlings of aerial images, region segmentation considering the rice clusters based on Voronoi Diagram, and uniformity index definition for evaluating the rice cluster distribution based on the variation coefficient. The results indicate the rice cluster recognition attains a high precision, with the precision, accuracy, recall, and F1-score of rice cluster recognition reaching > 95%, 97%, 97%, 95%, and 96%, respectively. The rice cluster location error is small and obeys the gamma (3.00, 0.54) distribution (mean error, 1.62 cm). The uniformity index is reasonable for evaluating the rice cluster distribution verified via simulation. As a whole process, the estimating method is sufficiently high accuracy with relative error less than 0.01% over the manual labeling method. Therefore, this method based on UAV images is feasible, convenient, technologically advanced, inexpensive, and highly precision for the estimation and evaluation of the rice cluster distribution uniformity. However, the evaluation application indicates that there is much room for improvement in terms of the uniformity of mechanized paddy field transplanting in South China.
format	article
author	Xiaohui Wang Qiyuan Tang Zhaozhong Chen Youyi Luo Hongyu Fu Xumeng Li
author_facet	Xiaohui Wang Qiyuan Tang Zhaozhong Chen Youyi Luo Hongyu Fu Xumeng Li
author_sort	Xiaohui Wang
title	Estimating and evaluating the rice cluster distribution uniformity with UAV-based images
title_short	Estimating and evaluating the rice cluster distribution uniformity with UAV-based images
title_full	Estimating and evaluating the rice cluster distribution uniformity with UAV-based images
title_fullStr	Estimating and evaluating the rice cluster distribution uniformity with UAV-based images
title_full_unstemmed	Estimating and evaluating the rice cluster distribution uniformity with UAV-based images
title_sort	estimating and evaluating the rice cluster distribution uniformity with uav-based images
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/5325ffc30c314994a05fbfc6bae22aa4
work_keys_str_mv	AT xiaohuiwang estimatingandevaluatingthericeclusterdistributionuniformitywithuavbasedimages AT qiyuantang estimatingandevaluatingthericeclusterdistributionuniformitywithuavbasedimages AT zhaozhongchen estimatingandevaluatingthericeclusterdistributionuniformitywithuavbasedimages AT youyiluo estimatingandevaluatingthericeclusterdistributionuniformitywithuavbasedimages AT hongyufu estimatingandevaluatingthericeclusterdistributionuniformitywithuavbasedimages AT xumengli estimatingandevaluatingthericeclusterdistributionuniformitywithuavbasedimages
_version_	1718442616106778624

Estimating and evaluating the rice cluster distribution uniformity with UAV-based images

Ejemplares similares