Strategies for monitoring within-field soybean yield using Sentinel-2 Vis-NIR-SWIR spectral bands and machine learning regression methods     

Crusiol  L. G.T.  Sun  Liang  Sibaldelli  R. N.R.  Junior  V. Felipe  Furlaneti  W. X.  Chen  R.  Sun  Z.  Wuyun  D.  Chen  Z.  Nanni  M. R.  Furlanetto  R. H.  Cezar  E.  Nepomuceno  A. L.  Farias  J. R.B. 《Precision Agriculture》2022,23(3):1093-1123

Soybean crop plays an important role in world food production and food security, and agricultural production should be increased accordingly to meet the global food demand. Satellite remote sensing data is considered a promising proxy for monitoring and predicting yield. This research aimed to evaluate strategies for monitoring within-field soybean yield using Sentinel-2 visible, near-infrared and shortwave infrared (Vis/NIR/SWIR) spectral bands and partial least squares regression (PLSR) and support vector regression (SVR) methods. Soybean yield maps (over 500 ha) were recorded by a combine harvester with a yield monitor in 15 fields (3 farms) in Paraná State, southern Brazil. Sentinel-2 images (spectral bands and 8 vegetation indices) across a cropping season were correlated to soybean yield. Information pooled across the cropping season presented better results compared to single images, with best performance of Vis/NIR/SWIR spectral bands under PLSR and SVR. At the grain filling stage, field-, farm- and global-based models were evaluated and presented similar trends compared to leaf-based hyperspectral reflectance collected at the Brazilian National Soybean Research Center. SVR outperformed PLSR, with a strong correlation between observed and predicted yield. For within-field soybean yield mapping, field-based SVR models (developed individually for each field) presented the highest accuracies. The results obtained demonstrate the possibility of developing within-field yield prediction models using Sentinel-2 Vis/NIR/SWIR bands through machine learning methods.

  相似文献   

NDVI variation according to the time of measurement,sampling size,positioning of sensor and water regime in different soybean cultivars     

Luis?Guilherme?Teixeira?Crusiol  Josirley?de?Fátima?Corrêa?Carvalho  Rubson?Natal?Ribeiro?Sibaldelli  Walkyria?Neiverth  Alexandre?do?Rio  Leonardo?Cesar?Ferreira  Sergio?de?Oliveira?Procópio  Liliane?Márcia?Mertz-Henning  Alexandre?Lima?Nepomuceno  Norman?Neumaier  José?Renato?Bou?as?FariasEmail author 《Precision Agriculture》2017,18(4):470-490

Although the information on the Normalized Difference Vegetation Index (NDVI) in plants under water deficit is often obtained from sensors attached to satellites, the increasing data acquisition with portable sensors has wide applicability in agricultural production because it is a fast, nondestructive method, and is less prone to interference problems. Thus, we carried out a set of experiments to investigate the influence of time, spatial plant arrangements, sampling size, height of the sensor and water regimes on NDVI readings in different soybean cultivars in greenhouse and field trials during the crop seasons 2011/12, 2012/13 and 2013/14. In experiments where plants were always evaluated under well-watered conditions, we observed that 9 a.m. was the most suitable time for NDVI readings regardless of the soybean cultivar, spatial arrangement or environment. Furthermore, there was no difference among NDVI readings in relation to the sampling size, regardless of the date or cultivar. We also observed that NDVI tended to decrease according to the higher height of the sensor in relation to the canopy top, with higher values tending to be at 0.8 m, but with no significant difference relative to 1.0 m—the height we adopted in our experiments. When different water regimes were induced under field conditions, NDVI readings measured at 9 a.m. by using a portable sensor were successful to differentiate soybean cultivars with contrasting responses to drought.  相似文献   

Vegetation indices and NIR-SWIR spectral bands as a phenotyping tool for water status determination in soybean     

Braga  P.  Crusiol  L. G. T.  Nanni  M. R.  Caranhato  A. L. H.  Fuhrmann  M. B.  Nepomuceno  A. L.  Neumaier  N.  Farias  J. R. B.  Koltun  A.  Gonçalves  L. S. A.  Mertz-Henning  L. M. 《Precision Agriculture》2021,22(1):249-266

Precision Agriculture - Drought is one of the main limiting factors of soybean production. The great deal of time and effort that current available phenotyping methods demand hampers the selection...  相似文献