The Global Wheat Head Detection(GWHD)dataset was created in 2020 and has assembled 193,634 labelled wheat heads from 4700 RGB images acquired from various acquisition platforms and 7 countries/institutions.With an ass...The Global Wheat Head Detection(GWHD)dataset was created in 2020 and has assembled 193,634 labelled wheat heads from 4700 RGB images acquired from various acquisition platforms and 7 countries/institutions.With an associated competition hosted in Kaggle,GWHD_2020 has successfully attracted attention from both the computer vision and agricultural science communities.From this first experience,a few avenues for improvements have been identified regarding data size,head diversity,and label reliability.To address these issues,the 2020 dataset has been reexamined,relabeled,and complemented by adding 1722 images from 5 additional countries,allowing for 81,553 additional wheat heads.We now release in 2021 a new version of the Global Wheat Head Detection dataset,which is bigger,more diverse,and less noisy than the GWHD_2020 version.展开更多
The detection of wheat heads in plant images is an important task for estimating pertinent wheat traits including head population density and head characteristics such as health,size,maturity stage,and the presence of...The detection of wheat heads in plant images is an important task for estimating pertinent wheat traits including head population density and head characteristics such as health,size,maturity stage,and the presence of awns.Several studies have developed methods for wheat head detection from high-resolution RGB imagery based on machine learning algorithms.However,these methods have generally been calibrated and validated on limited datasets.High variability in observational conditions,genotypic differences,development stages,and head orientation makes wheat head detection a challenge for computer vision.Further,possible blurring due to motion or wind and overlap between heads for dense populations make this task even more complex.Through a joint international collaborative effort,we have built a large,diverse,and well-labelled dataset of wheat images,called the Global Wheat Head Detection(GWHD)dataset.It contains 4700 high-resolution RGB images and 190000 labelled wheat heads collected from several countries around the world at different growth stages with a wide range of genotypes.Guidelines for image acquisition,associating minimum metadata to respect FAIR principles,and consistent head labelling methods are proposed when developing new head detection datasets.The GWHD dataset is publicly available at http://www.global-wheat.com/and aimed at developing and benchmarking methods for wheat head detection.展开更多
基金the French National Research Agency under the Investments for the Future Program,referred as ANR-16-CONV-0004 PIA#Digitag.Institut Convergences Agriculture Numérique,Hiphen supported the organization of the competition.Japan:Kubota supported the organization of the competi-tion.Australia:Grains Research and Development Corpora-tion(UOQ2002-008RTX machine learning applied to high-throughput feature extraction from imagery to map spatial variability and UOQ2003-011RTX INVITA-a technology and analytics platform for improving variety selection)sup-ported competition.
文摘The Global Wheat Head Detection(GWHD)dataset was created in 2020 and has assembled 193,634 labelled wheat heads from 4700 RGB images acquired from various acquisition platforms and 7 countries/institutions.With an associated competition hosted in Kaggle,GWHD_2020 has successfully attracted attention from both the computer vision and agricultural science communities.From this first experience,a few avenues for improvements have been identified regarding data size,head diversity,and label reliability.To address these issues,the 2020 dataset has been reexamined,relabeled,and complemented by adding 1722 images from 5 additional countries,allowing for 81,553 additional wheat heads.We now release in 2021 a new version of the Global Wheat Head Detection dataset,which is bigger,more diverse,and less noisy than the GWHD_2020 version.
基金The French team received support from ANRT for the CIFRE grant of Etienne David,cofunded by Arvalis.The study was partly supported by several projects including ANR PHENOME,ANR BREEDWHEAT,CASDAR LITERAL,and FSOV“Plastix”.Many thanks are due to the people who annotated the French datasets,including Frederic Venault,Xiuliang Jin,Mario Serouard,Ilias Sarbout,Carole Gigot,Eloïse Issert,and Elise Lepage.The Japanese team received support from JST CREST(Grant Numbers JPMJCR16O3,JPMJCR16O2,and JPMJCR1512)and MAFF Smart-Breeding System for Innovative Agriculture(BAC1003),Japan.Many thanks are due to the people who annotated the Japanese dataset,including Kozue Wada,Masanori Ishii,Ryuuichi Kanzaki,Sayoko Ishibashi,and Sumiko Kaneko.The Canadian team received funding from the Plant Phenotyping and Imaging Research Center through a grant from the Canada First Research Excellence Fund.Many thanks are due to Steve Shirtliffe,Scott Noble,Tyrone Keep,Keith Halco,and Craig Gavelin for managing the field site and collecting images.Rothamsted Research received support from the Biotechnology and Biological Sciences Research Council(BBSRC)of the United Kingdom as part of the Designing Future Wheat(BB/P016855/1)project.We are also thankful to Prof.MalcolmJ.Hawkesford,who leads the DFWproject and Dr.Nicolas Virlet for conducting the experiment at Rothamsted Research.The Gatton,Australia dataset was collected on a field trial conducted by CSIRO and UQ,with trial conduct and measurements partly funded by the Grains Research and Development Corporation(GRDC)in project CSP00179.A new GRDC project involves several of the authors and supports their contribution to this paper.The dataset collected in China was supported by the Program for High-Level Talents Introduction of Nanjing Agricultural University(440—804005).Many thanks are due to Jie Zhou and many volunteers from Nanjing Agricultural University to accomplish the annotation.The dataset collection at ETHZ was supported by Prof.AchimWalter,who leads the Crop Science group.Many thanks are due to Kevin Keller for the initial preparation of the ETHZ dataset and Lara Wyser,Ramon Winterberg,Damian Käch,Marius Hodel,and Mario Serouard(INRAE)for the annotation of the ETHZ dataset and to Brigita Herzog and Hansueli Zellweger for crop husbandry.
文摘The detection of wheat heads in plant images is an important task for estimating pertinent wheat traits including head population density and head characteristics such as health,size,maturity stage,and the presence of awns.Several studies have developed methods for wheat head detection from high-resolution RGB imagery based on machine learning algorithms.However,these methods have generally been calibrated and validated on limited datasets.High variability in observational conditions,genotypic differences,development stages,and head orientation makes wheat head detection a challenge for computer vision.Further,possible blurring due to motion or wind and overlap between heads for dense populations make this task even more complex.Through a joint international collaborative effort,we have built a large,diverse,and well-labelled dataset of wheat images,called the Global Wheat Head Detection(GWHD)dataset.It contains 4700 high-resolution RGB images and 190000 labelled wheat heads collected from several countries around the world at different growth stages with a wide range of genotypes.Guidelines for image acquisition,associating minimum metadata to respect FAIR principles,and consistent head labelling methods are proposed when developing new head detection datasets.The GWHD dataset is publicly available at http://www.global-wheat.com/and aimed at developing and benchmarking methods for wheat head detection.
