Global Wheat Head Detection Challenges:Winning Models and Application for Head Counting

作     者：Etienne David Franklin Ogidi Daniel Smith Scott Chapman Benoit de Solan Wei Guo Frederic Baret Ian Stavness Etienne David;Franklin Ogidi;Daniel Smith;Scott Chapman;Benoit de Solan;Wei Guo;Frederic Baret;Ian Stavness

作者机构：UMR 1114 EMMAHINRAEAvignonFrance Arvalis–Institut du VégétalParisFrance Department of Computer ScienceUniversity of SaskatchewanSaskatoonCanada School of Food and Agricultural SciencesUniversity of QueenslandBrisbaneAustralia Graduate School of Agricultural and Life SciencesThe University of TokyoTokyoJapan 

出 版 物：《Plant Phenomics》 (植物表型组学（英文）)

年 卷 期：2023年第5卷第3期

页      面：460-473页

核心收录：

学科分类：0710[理学-生物学] 09[农学] 0816[工学-测绘科学与技术] 0901[农学-作物学] 

基　　金：support from ANRT for the CIFRE grant of E.D.,cofunded by Arvalis partly supported by several projects,including:Canada:The Canada First Research Excellence Fund and the Global Institute Food Security,University of Saskatchewan supported the organization of the competition. rance:PIA#Digitag Institut Convergences Agriculture Numérique,Hiphen sup-ported the organization of the competition and the Agence Nationale de la Recherche projects ANR-11-INBS-0012(Phenome) Japan:Kubota supported the organization of the competition Australia:Grains Research and Development Corporation(UOQ2002-008RTX Machine learning applied to high-throughput feature extraction from imagery to map spatial variability and UOQ2003-011RTX INVITA-A technol-ogy and analytics platform for improving variety selection)supported competition and data provision/discussions. 

主　　题：Wheat specialized competition 

摘      要：Data competitions have become a popular approach to crowdsource new data analysis methods for general and specialized data science problems.Data competitions have a rich history in plant phenotyping,and new outdoor field datasets have the potential to embrace solutions across research and commercial applications.We developed the Global Wheat Challenge as a generalization competition in 2020 and 2021 to find more robust solutions for wheat head detection using field images from different regions.We analyze the winning challenge solutions in terms of their robustness when applied to new datasets.We found that the design of the competition had an influence on the selection of winning solutions and provide recommendations for future competitions to encourage the selection of more robust solutions.