DOI：10.31220/agriRxiv.2022.00155

Deep convolutional neural networks exploit high spatial and temporal resolution aerial imagery to predict key traits in miscanthus.

•  

  Abstract: Abstract Miscanthus is one of the most promising perennial crops for bioenergy production, with high yield potential and a low environmental footprint. The increasing interest in this crop requires accelerated selection and the development of new screening techniques. New analytical methods that are more accurate and less labor intensive are needed to better characterize the effects of genetics and the environment on key traits under field conditions. We used persistent multispectral and photogrammetric UAV time-series imagery collected 10 times in the season, together with ground-truth data over thousands of miscanthus genotypes to determine flowering time, culm length, and biomass yield traits. We compared the performance of Convolutional Neural Network (CNN) architectures that used image data from single dates (2D-spatial) versus the integration of multiple dates by (3D-spatio-temporal) architectures. The ability of UAV-based remote sensing to rapidly and non-destructively assess large-scale genetic variation in flowering time, height and biomass production was improved through use of 3D-spatio-temporal CNN architectures versus 2D-spatial CNN architectures. The performance gains of the best 3D-spatio-temporal analyses compared to the best 2D-spatial architectures manifested in up to: 23% improvements in R 2 , 17% reductions in RMSE, and 20% reductions in MAE. The integration of photogrammetric and spectral features with 3D- architecture was key to the improved assessment of all traits. In conclusion, this study demonstrates that the integration of high spatiotemporal resolution UAV imagery with 3D-CNNs enables more accurate monitoring of the dynamics of key crop phenological and yield-related traits. This is especially valuable in highly productive, perennial grass crops such as miscanthus, where in-field phenotyping is especially challenging and traditionally limits the rate of crop improvement through breeding.

  Key words: key; 3d; miscanthus; the integration of; traits; 2

  提交时间：2022-01-01

  版权声明：作者本人独立拥有该论文的版权，预印本系统仅拥有论文的永久保存权利。任何人未经允许不得重复使用。
• html

• 图表

• Sebastián Varela, Andrew D. B. Leake. Implementing spatio-temporal 3D-convolution neural networks and UAV time series imagery to better predict lodging damage in sorghum.. 2022. doi: 10.31220/agriRxiv.2022.00120

  Ines Adriaens, W. Ouweltjes, Matti Pastell, C. Kamphui. Detecting dairy cows' lying behavior using noisy 3D ultra-wide band positioning data.. 2022. doi: 10.31220/agriRxiv.2022.00132

  Babar Hussain, Bala Anı Akpınar, Michaël Alaux, A. M. Algharib, Deepmala Sehgal, Zulfiqar Ali, R. Appels, G. I. Aradottir, Jacqueline Batley, Arnaud Bellec, Alison R. Bentley, H Busra Cagirici, L. Cattivelli, Frédéric Choulet, James Cockram, Flora Desiderio, Pierre Devaux, Münevver Doğramacı, Gabriel Dorado, Susanne Dreisigacker, Dave Edwards, K. El-Hassouni, Kellye Eversole, Tzion Fahima, Melania Figueroa, Sergio Gálvez, Kulvinder S. Gill, Liubov Govta, Alvina Gul, Goetz Hensel, Pilar Hernández, L. C. Herrera, Amir M. H. Ibrahim, Benjamin Kilian, Viktor Korzun, Tamar Krugman, Yinghui Li, Shuyu Liu, Amer F. Mahmoud, Alexey Morgounov, Tuğdem Muslu, Faiza Naseer, Frank Ordon, Etienne Paux, Dragan Perović, Gadi V. P. Reddy, Jochen C. Reif, Matthew Reynolds, Rajib Roychowdhury, Jackie C. Rudd, Taner Z. Sen, Sivakumar Sukumaran, Vijay K. Tiwari, Naimat Ullah, Turgay Ünver, Selami Yazar, Hikmet Buda. Wheat genomics and breeding: bridging the gap.. 2021. doi: 10.31220/agriRxiv.2021.00039

  Midian Nascimento dos Santos, Daniel Rothschild, Tina M. Widowski, Shai Barbut, E. Kiarie, I. B. Mandell, Michele T. Guerin, A. Michelle Edwards, Stephanie Torre. In pursuit of a better broiler: carcass traits and muscle myopathies in conventional and slower-growing strains of broiler chickens.. 2021. doi: 10.31220/agriRxiv.2021.00037

  Addisu Getahu. Estimating genetic variability, gene effect, heritability, and genetic advance of yield component agronomic traits: towards the improvement of peanut agronomic traits.. 2021. doi: 10.31220/agriRxiv.2021.00058

  Estelle Leroux, I. Llach, Gaelle Besche, Eliel González García, D. Montier, D. Montier, Pierre Marie Bouque. Growth and behaviour of post-weaned Mérinos d'Arles ewe lambs, as measured by a Walk-over-Weighing system under Mediterranean grazing conditions.. 2021. doi: 10.31220/agriRxiv.2021.00109

  Habte Jifar, Zerihun Tadele, Kassahun Tesfaye, Kebebew Assefa, Kifle Dagne, Alemayehu Teressa Negawo, Michael Blümmel, Chris S. Jone. Exploring the potential to capture variation in grain-straw yield and straw quality traits of tef [Eragrostis tef (Zucc.) Trotter] varieties.. 2021. doi: 10.31220/agriRxiv.2021.00078

  Tashin Ahmed, Chowdhury Rafeed Rahman, Md. Faysal Mahmud Abi. Rice grain disease identification using dual phase convolutional neural network based system aimed at small dataset.. 2021. doi: 10.31220/agriRxiv.2021.00062

  Sospeter Gachamba, Yao‐Wu Xing, Kelsey F. Andersen, Karen A. Garrett, Douglas Watuku Miano, A. W. Mwang’ombe, Kalpana Sharm. Epidemic networks and potential sources of bacterial wilt infection in a potato seed network in Kenya.. 2022. doi: 10.31220/agriRxiv.2022.00117

  Nuriel Shalom Mor, Tigabo Asras, Eli Gal, Tesfahon Demasia, Ezra Tarab, Nathaniel Ezekiel, Osher Nikapros, Oshri Semimufar, Eva Gladky, Maria Karpenko, Daniel Sason, Д. Н. Маслов, Omri Mo. Wine quality and type prediction from physicochemical properties using neural networks for machine learning: a free software for winemakers and customers.. 2022. doi: 10.31220/agriRxiv.2022.00125

• 序号

  提交日期

  编号

  操作

• 公开评论  匿名评论  仅发给作者 提交评论