Legume content estimation from UAV image in grass-legume meadows: comparison methods based on the UAV coverage vs. field biomass

Abstract Legume content (LC) in grass-legume mixtures is important for assessing forage quality and optimizing fertilizer application in meadow fields. This study focuses on differences in LC measurements obtained from unmanned aerial vehicle (UAV) images and ground surveys based on dry matter asses...

Full description

Saved in:
Bibliographic Details
Main Authors: Kensuke Kawamura, Tsuneki Tanaka, Taisuke Yasuda, Shoji Okoshi, Masaaki Hanada, Kazuya Doi, Toshiya Saigusa, Takanori Yagi, Kenji Sudo, Kenji Okumura, Jihyun Lim
Format: Article
Language:English
Published: Nature Portfolio 2024-12-01
Series:Scientific Reports
Online Access:https://doi.org/10.1038/s41598-024-82055-w
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850282520878252032
author Kensuke Kawamura
Tsuneki Tanaka
Taisuke Yasuda
Shoji Okoshi
Masaaki Hanada
Kazuya Doi
Toshiya Saigusa
Takanori Yagi
Kenji Sudo
Kenji Okumura
Jihyun Lim
author_facet Kensuke Kawamura
Tsuneki Tanaka
Taisuke Yasuda
Shoji Okoshi
Masaaki Hanada
Kazuya Doi
Toshiya Saigusa
Takanori Yagi
Kenji Sudo
Kenji Okumura
Jihyun Lim
author_sort Kensuke Kawamura
collection DOAJ
description Abstract Legume content (LC) in grass-legume mixtures is important for assessing forage quality and optimizing fertilizer application in meadow fields. This study focuses on differences in LC measurements obtained from unmanned aerial vehicle (UAV) images and ground surveys based on dry matter assessments in seven meadow fields in Hokkaido, Japan. We propose a UAV-based LC (LCUAV) estimation and mapping method using a land cover map from a simple linear iterative clustering (SLIC) algorithm and a random forest (RF) classifier. The SLIC-RF classification achieved a high accuracy level for four different ground cover types (grasses, legumes, weeds, and background) in seven distinct meadows with an overall accuracy of 91.4% and an F score of 91.5%. By applying SLIC-RF to eliminate plots with low classification accuracy, we demonstrate the necessity of achieving a minimum classification accuracy of 0.82 for precise LC estimation. A non-linear relationship was revealed between the LCUAV and LCBM influenced by surface sward height (SSH, height of plant canopy). The results indicate a higher accuracy of the LCBM estimation when SSH levels were lower, particularly when recommending SSH levels below 40 cm for optimal LCBM estimation. This highlights the effectiveness of UAV-based remote sensing for assessing early growth or grazing in pastures with low SSH.
format Article
id doaj-art-920be330c7af4e8795db5cbf8bb80d56
institution OA Journals
issn 2045-2322
language English
publishDate 2024-12-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-920be330c7af4e8795db5cbf8bb80d562025-08-20T01:47:58ZengNature PortfolioScientific Reports2045-23222024-12-0114111310.1038/s41598-024-82055-wLegume content estimation from UAV image in grass-legume meadows: comparison methods based on the UAV coverage vs. field biomassKensuke Kawamura0Tsuneki Tanaka1Taisuke Yasuda2Shoji Okoshi3Masaaki Hanada4Kazuya Doi5Toshiya Saigusa6Takanori Yagi7Kenji Sudo8Kenji Okumura9Jihyun Lim10Obihiro University of Agriculture and Veterinary MedicineDairy Research Center, Hokkaido Research Organization (HRO)Mount Fuji Research Institute, Yamanashi Prefectural GovernmentObihiro University of Agriculture and Veterinary MedicineObihiro University of Agriculture and Veterinary MedicineRakuno Gakuen UniversityRakuno Gakuen UniversityHokkaido Agricultural Research Center, National Agriculture and Food Research Organization (NARO)Hokkaido Agricultural Research Center, National Agriculture and Food Research Organization (NARO)Hokkaido Agricultural Research Center, National Agriculture and Food Research Organization (NARO)Institute for Agro-Environmental Sciences, National Agriculture and Food Research Organization (NARO)Abstract Legume content (LC) in grass-legume mixtures is important for assessing forage quality and optimizing fertilizer application in meadow fields. This study focuses on differences in LC measurements obtained from unmanned aerial vehicle (UAV) images and ground surveys based on dry matter assessments in seven meadow fields in Hokkaido, Japan. We propose a UAV-based LC (LCUAV) estimation and mapping method using a land cover map from a simple linear iterative clustering (SLIC) algorithm and a random forest (RF) classifier. The SLIC-RF classification achieved a high accuracy level for four different ground cover types (grasses, legumes, weeds, and background) in seven distinct meadows with an overall accuracy of 91.4% and an F score of 91.5%. By applying SLIC-RF to eliminate plots with low classification accuracy, we demonstrate the necessity of achieving a minimum classification accuracy of 0.82 for precise LC estimation. A non-linear relationship was revealed between the LCUAV and LCBM influenced by surface sward height (SSH, height of plant canopy). The results indicate a higher accuracy of the LCBM estimation when SSH levels were lower, particularly when recommending SSH levels below 40 cm for optimal LCBM estimation. This highlights the effectiveness of UAV-based remote sensing for assessing early growth or grazing in pastures with low SSH.https://doi.org/10.1038/s41598-024-82055-w
spellingShingle Kensuke Kawamura
Tsuneki Tanaka
Taisuke Yasuda
Shoji Okoshi
Masaaki Hanada
Kazuya Doi
Toshiya Saigusa
Takanori Yagi
Kenji Sudo
Kenji Okumura
Jihyun Lim
Legume content estimation from UAV image in grass-legume meadows: comparison methods based on the UAV coverage vs. field biomass
Scientific Reports
title Legume content estimation from UAV image in grass-legume meadows: comparison methods based on the UAV coverage vs. field biomass
title_full Legume content estimation from UAV image in grass-legume meadows: comparison methods based on the UAV coverage vs. field biomass
title_fullStr Legume content estimation from UAV image in grass-legume meadows: comparison methods based on the UAV coverage vs. field biomass
title_full_unstemmed Legume content estimation from UAV image in grass-legume meadows: comparison methods based on the UAV coverage vs. field biomass
title_short Legume content estimation from UAV image in grass-legume meadows: comparison methods based on the UAV coverage vs. field biomass
title_sort legume content estimation from uav image in grass legume meadows comparison methods based on the uav coverage vs field biomass
url https://doi.org/10.1038/s41598-024-82055-w
work_keys_str_mv AT kensukekawamura legumecontentestimationfromuavimageingrasslegumemeadowscomparisonmethodsbasedontheuavcoveragevsfieldbiomass
AT tsunekitanaka legumecontentestimationfromuavimageingrasslegumemeadowscomparisonmethodsbasedontheuavcoveragevsfieldbiomass
AT taisukeyasuda legumecontentestimationfromuavimageingrasslegumemeadowscomparisonmethodsbasedontheuavcoveragevsfieldbiomass
AT shojiokoshi legumecontentestimationfromuavimageingrasslegumemeadowscomparisonmethodsbasedontheuavcoveragevsfieldbiomass
AT masaakihanada legumecontentestimationfromuavimageingrasslegumemeadowscomparisonmethodsbasedontheuavcoveragevsfieldbiomass
AT kazuyadoi legumecontentestimationfromuavimageingrasslegumemeadowscomparisonmethodsbasedontheuavcoveragevsfieldbiomass
AT toshiyasaigusa legumecontentestimationfromuavimageingrasslegumemeadowscomparisonmethodsbasedontheuavcoveragevsfieldbiomass
AT takanoriyagi legumecontentestimationfromuavimageingrasslegumemeadowscomparisonmethodsbasedontheuavcoveragevsfieldbiomass
AT kenjisudo legumecontentestimationfromuavimageingrasslegumemeadowscomparisonmethodsbasedontheuavcoveragevsfieldbiomass
AT kenjiokumura legumecontentestimationfromuavimageingrasslegumemeadowscomparisonmethodsbasedontheuavcoveragevsfieldbiomass
AT jihyunlim legumecontentestimationfromuavimageingrasslegumemeadowscomparisonmethodsbasedontheuavcoveragevsfieldbiomass