scholarly journals Assessment of Ensemble Learning to Predict Wheat Grain Yield Based on UAV-Multispectral Reflectance

2021 ◽  
Vol 13 (12) ◽  
pp. 2338
Author(s):  
Shuaipeng Fei ◽  
Muhammad Adeel Hassan ◽  
Zhonghu He ◽  
Zhen Chen ◽  
Meiyan Shu ◽  
...  
Keyword(s):  
Grain Yield ◽  
Ensemble Learning ◽  
Complex Traits ◽  
Multispectral Imaging ◽  
Grain Filling ◽  
Support Vector ◽  
Growth Stages ◽  
Yield Prediction ◽  
Multispectral Data ◽  
Irrigation Treatments

Grain yield is increasingly affected by climate factors such as drought and heat. To develop resilient and high-yielding cultivars, high-throughput phenotyping (HTP) techniques are essential for precise decisions in wheat breeding. The ability of unmanned aerial vehicle (UAV)-based multispectral imaging and ensemble learning methods to increase the accuracy of grain yield prediction in practical breeding work is evaluated in this study. For this, 211 winter wheat genotypes were planted under full and limited irrigation treatments, and multispectral data were collected at heading, flowering, early grain filling (EGF), and mid-grain filling (MGF) stages. Twenty multispectral vegetation indices (VIs) were estimated, and VIs with heritability greater than 0.5 were selected to evaluate the models across the growth stages under both irrigation treatments. A framework for ensemble learning was developed by combining multiple base models such as random forest (RF), support vector machine (SVM), Gaussian process (GP), and ridge regression (RR). The R2 values between VIs and grain yield between for individual base models were ranged from 0.468 to 0.580 and 0.537 to 0.598 for grain yield prediction in full and limited irrigation treatments across growth stages, respectively. The prediction results of ensemble models were ranged from 0.491 to 0.616 and 0.560 to 0.616 under full and limited irrigation treatments respectively, and were higher than that of the corresponding base learners. Moreover, the grain yield prediction results were observed high at mid grain filling stage under both full (R2 = 0.625) and limited (R2 = 0.628) irrigation treatments through ensemble learning based stacking of four base learners. Further improvements in ensemble learning models can accelerate the use of UAV-based multispectral data for accurate predictions of complex traits like grain yield in wheat.

scholarly journals Yield Prediction by Machine Learning From UAS-Based Multi-Sensor Data Fusion in Soybean

2020 ◽  
Author(s):  
Monica Herrero-Huerta ◽  
Pablo Rodriguez-Gonzalvez ◽  
Katy M. Rainey
Keyword(s):  
Machine Learning ◽  
Grain Yield ◽  
Growing Season ◽  
Sensor Data ◽  
Growth Stages ◽  
Yield Prediction ◽  
Proactive Management ◽  
Combining Data ◽  
Extreme Gradient Boosting ◽  
System Solution

Abstract Background Nowadays, automated phenotyping of plants is essential for precise and cost-effective improvement in the efficiency of crop genetics. In recent years, machine learning (ML) techniques have shown great success in the classification and modelling of crop parameters. In this research, we consider the capability of ML to perform grain yield prediction in soybeans by combining data from different optical sensors via RF (Random Forest) and XGBoost (eXtreme Gradient Boosting). During the 2018 growing season, a panel of 382 soybean recombinant inbred lines were evaluated in a yield trial at the Agronomy Center for Research and Education (ACRE) in West Lafayette (Indiana, USA). Images were acquired by the Parrot Sequoia Multispectral Sensor and the S.O.D.A. compact digital camera on board a senseFly eBee UAS (Unnamed Aircraft System) solution at R4 and early R5 growth stages. Next, a standard photogrammetric pipeline was carried out by SfM (Structure from Motion). Multispectral imagery serves to analyse the spectral response of the soybean end-member in 2D. In addition, RGB images were used to reconstruct the study area in 3D, evaluating the physiological growth dynamics per plot via height variations and crop volume estimations. As ground truth, destructive grain yield measurements were taken at the end of the growing season. Results Algorithms and feature extraction techniques were combined to develop a regression model to predict final yield from imagery, achieving an accuracy of over 90.72% by RF and 91.36% by XGBoost. Conclusions Results provide practical information for the selection of phenotypes for breeding coming from UAS data as a decision support tool, affording constant operational improvement and proactive management for high spatial precision.

Plant population, shading and thinning studies in wheat

1971 ◽  
Vol 77 (3) ◽  
pp. 453-461 ◽  
Author(s):  
R. W. Willey ◽  
R. Holliday
Keyword(s):  
Grain Yield ◽  
Dry Matter ◽  
Unit Area ◽  
Grain Filling ◽  
Plant Population ◽  
Growth Stages ◽  
Response Curves ◽  
Higher Plant ◽  
Plant Populations ◽  
Ear Development

SUMMARYThree wheat experiments are described in which a range of plant populations were shaded during different periods of development; in two of the experiments plant thinning was also carried out at a number of growth stages. Shading during the period of ear development caused an appreciable decrease in grain yield by decreasing the number of grains per ear. Shading during the grain filling period also reduced grain yield, this being brought about by decreased grain size. Thus in contrast to the barley experiments reported earlier (Willey & Holhday, 1971), these particular results gave no indication of a potential surplus of carbohydrate for grain filling and an associated limited ear capacity. However, when plant thinning was carried out at anthesis to make more carbohydrate available for grain filling in the remaining ears, grain yield per ear did not increase. It is argued, therefore, that grain yield probably was determined at least partly by a limited ear capacity. Plant thinning at earlier stages showed how the development of competition during the ear development period progressively reduced the potential capacity of the ear; the greater competition of higher plant populations accelerated this reduction in ear potential.From an examination of the effects of plant population, it is suggested that the number of grains per ear is the component having greatest influence on the decline in grain yield at above-optimum populations. The possible importance of the number of grains per unit area as an indicator of ear capacity on an area basis, and as a determinant of grain yield per unit area, is emphasized. A close relationship between grain yield per unit area and number of grains per unit area is illustrated for a number of plant-population response curves, and it is suggested that the decrease in grain yield at high populations is probably determined by a decrease in the number of grains per unit area. Evidence is presented to substantiate the idea put forward in the barley paper that this decrease in the number of grains per unit area may be attributable more to a lower production of total dry matter by the high populations during the later stages of ear development, than to an unfavourable partitioning of such dry matter between the ear and the rest of the plant.

scholarly journals Improvement of Wheat Grain Yield Prediction Model Performance Based on Stacking Technique

2021 ◽  
Vol 11 (24) ◽  
pp. 12164
Author(s):  
Changchun Li ◽  
Yilin Wang ◽  
Chunyan Ma ◽  
Weinan Chen ◽  
Yacong Li ◽  
...  
Keyword(s):  
Prediction Model ◽  
Grain Yield ◽  
Growth Stage ◽  
Prediction Models ◽  
Vegetation Indices ◽  
Model Performance ◽  
Grain Filling ◽  
Growth Stages ◽  
Wheat Grain ◽  
Area Index

Crop growth and development is a dynamic and complex process, and the essence of yield formation is the continuous accumulation of photosynthetic products from multiple fertility stages. In this study, a new stacking method for integrating multiple growth stages information was proposed to improve the performance of the winter wheat grain yield (GY) prediction model. For this purpose, crop canopy hyperspectral reflectance and leaf area index (LAI) data were obtained at the jointing, flagging, anthesis and grain filling stages. In this case, 15 vegetation indices and LAI were used as input features of the elastic network to construct GY prediction models for single growth stage. Based on Stacking technique, the GY prediction results of four single growth stages were integrated to construct the ensemble learning framework. The results showed that vegetation indices coupled LAI could effectively overcome the spectral saturation phenomenon, the validated R2 of each growth stage was improved by 10%, 22.5%, 3.6% and 10%, respectively. The stacking method provided more stable information with higher prediction accuracy than the individual fertility results (R2 = 0.74), and the R2 of the model validation phase improved by 236%, 51%, 27.6%, and 12.1%, respectively. The study can provide a reference for GY prediction of other crops.

scholarly journals Principal variable selection to explain grain yield variation in winter wheat from features extracted from UAV imagery

Plant Methods ◽  
2019 ◽  
Vol 15 (1) ◽  
Author(s):  
Jiating Li ◽  
Arun-Narenthiran Veeranampalayam-Sivakumar ◽  
Madhav Bhatta ◽  
Nicholas D. Garst ◽  
Hannah Stoll ◽  
...  
Keyword(s):  
Random Forest ◽  
Winter Wheat ◽  
Grain Yield ◽  
Plant Height ◽  
Vegetation Indices ◽  
Growing Season ◽  
Growth Stages ◽  
Yield Prediction ◽  
Secondary Traits ◽  
Parametric Algorithm

Abstract Background Automated phenotyping technologies are continually advancing the breeding process. However, collecting various secondary traits throughout the growing season and processing massive amounts of data still take great efforts and time. Selecting a minimum number of secondary traits that have the maximum predictive power has the potential to reduce phenotyping efforts. The objective of this study was to select principal features extracted from UAV imagery and critical growth stages that contributed the most in explaining winter wheat grain yield. Five dates of multispectral images and seven dates of RGB images were collected by a UAV system during the spring growing season in 2018. Two classes of features (variables), totaling to 172 variables, were extracted for each plot from the vegetation index and plant height maps, including pixel statistics and dynamic growth rates. A parametric algorithm, LASSO regression (the least angle and shrinkage selection operator), and a non-parametric algorithm, random forest, were applied for variable selection. The regression coefficients estimated by LASSO and the permutation importance scores provided by random forest were used to determine the ten most important variables influencing grain yield from each algorithm. Results Both selection algorithms assigned the highest importance score to the variables related with plant height around the grain filling stage. Some vegetation indices related variables were also selected by the algorithms mainly at earlier to mid growth stages and during the senescence. Compared with the yield prediction using all 172 variables derived from measured phenotypes, using the selected variables performed comparable or even better. We also noticed that the prediction accuracy on the adapted NE lines (r = 0.58–0.81) was higher than the other lines (r = 0.21–0.59) included in this study with different genetic backgrounds. Conclusions With the ultra-high resolution plot imagery obtained by the UAS-based phenotyping we are now able to derive more features, such as the variation of plant height or vegetation indices within a plot other than just an averaged number, that are potentially very useful for the breeding purpose. However, too many features or variables can be derived in this way. The promising results from this study suggests that the selected set from those variables can have comparable prediction accuracies on the grain yield prediction than the full set of them but possibly resulting in a better allocation of efforts and resources on phenotypic data collection and processing.

scholarly journals Evaluation of rice grain yield and yield components of Nona Bokra chromosome segment substitution lines with the genetic background of Koshihikari, in a saline paddy field

AoB Plants ◽  
2019 ◽  
Vol 11 (5) ◽  
Author(s):  
Shiro Mitsuya ◽  
Norifumi Murakami ◽  
Tadashi Sato ◽  
Tomohiko Kazama ◽  
Kinya Toriyama ◽  
...  
Keyword(s):  
Grain Yield ◽  
Paddy Field ◽  
Yield Components ◽  
Genetic Background ◽  
Grain Filling ◽  
Growth Stages ◽  
Substitution Lines ◽  
Chromosome Segment Substitution Lines ◽  
Yield And Yield Components ◽  
Segment Substitution

Abstract The ability to tolerate salt differs with the growth stages of rice and thus the yield components that are determined during various growth stages, are differentially affected by salt stress. In this study, we utilized chromosome segment substitution lines (CSSLs) from Nona Bokra, a salt-tolerant indica landrace, with the genetic background of Koshihikari, a salt-susceptible japonica variety. These were screened to find superior CSSLs under long-term saline conditions that showed higher grain yield and yield components in comparison to Koshihikari. One-month-old seedlings were transplanted into a paddy field without salinity. These were allowed to establish for 1 month further, then the field was flooded, with saline water maintained at 7.41 dS m−1 salinity until harvest. The experiments were performed twice, once in 2015 and a targeted study in 2016. Salt tolerance of growth and reproductive stage parameters was evaluated as the Salt Effect Index (SEI) which was computed as the difference in each parameter within each line between control and saline conditions. All CSSLs and Koshihikari showed a decrease in grain yield and yield components except panicle number under salinity. SL538 showed a higher SEI for grain yield compared with Koshihikari under salinity throughout the two experiments. This was attributed to the retained grain filling and harvest index, yet the mechanism was not due to maintaining Na+, Cl− and K+ homeostasis. Few other CSSLs showed greater SEI for grain weight under salinity compared with Koshihikari, which might be related to low concentration of Na+ in leaves and panicles. These data indicate that substitution of different Nona Bokra chromosome segments independently contributed to the maintenance of grain filling and grain weight of Koshihikari under saline conditions.

scholarly journals Modelling the effects of post-heading heat stress on biomass partitioning, and grain number and weight of wheat

2020 ◽  
Vol 71 (19) ◽  
pp. 6015-6031
Author(s):  
Bing Liu ◽  
Leilei Liu ◽  
Senthold Asseng ◽  
Dongzheng Zhang ◽  
Wei Ma ◽  
...  
Keyword(s):  
Heat Stress ◽  
Grain Yield ◽  
Crop Production ◽  
Grain Filling ◽  
Biomass Partitioning ◽  
Grain Weight ◽  
Growth Stages ◽  
Grain Number ◽  
Good Ability ◽  
Improved Model

Abstract Grain yield of wheat and its components are very sensitive to heat stress at the critical growth stages of anthesis and grain filling. We observed negative impacts of heat stress on biomass partitioning and grain growth in environment-controlled phytotron experiments over 4 years, and we quantified relationships between the stress and grain number and potential grain weight at anthesis and during grain filling using process-based heat stress routines. These relationships included reduced grain set under stress at anthesis and decreased potential grain weight under stress during early grain filling. Biomass partitioning to stems and spikes was modified under heat stress based on a source–sink relationship. The integration of our process-based stress routines into the original WheatGrow model significantly enhanced the predictions of the biomass dynamics of the stems and spikes, the grain yield, and the yield components under heat stress. Compared to the original model, the improved version decreased the simulation errors for grain yield, grain number, and grain weight by 73%, 48%, and 49%, respectively, in an evaluation using independent data under heat stress in the phytotron conditions. When tested with data obtained under field conditions, the improved model showed a good ability to reproduce the decreasing dynamics of grain yield and its components with increasing post-anthesis temperatures. Sensitivity analysis showed that the improved model was able to reproduce the responses to various observed heat-stress treatments. These improvements to the crop model will be of significant importance for assessing the effects on crop production of projected increases in heat-stress events under future climate scenarios.

scholarly journals Evaluation of genotypic response of kabuli chickpea (cicer arietinum l.) cultivars to irrigation regimes in northwest of iran

2014 ◽  
Vol 60 (1) ◽  
pp. 22-30
Author(s):  
Mokhtar Pasandi ◽  
Mohsen Janmohammadi ◽  
Rahmatollah Karimizadeh
Keyword(s):  
Grain Yield ◽  
Cicer Arietinum ◽  
Root Zone ◽  
Grain Filling ◽  
Growth Stages ◽  
Full Irrigation ◽  
Cicer Arietinum L ◽  
Irrigation Regimes ◽  
Significant Difference ◽  
Semi Arid

Abstract Water deficiency is commonly the most important yield -restraining factor in semi-arid and Mediterranean environments. Chickpea (Cicer arietinum L.), which is one the main legume crops of the region, often experiences terminal drought. To investigate the response of chickpea genotypes to different irrigation levels, experiments were conducted in Maragheh, Northwest Iran. Three levels of irrigation including zero (rain-fed condition), full irrigation (enough water to fill the root zone profile) and two supplement irrigations (SIs) during flowering and grain filling stages were evaluated over 2013 growing season. Results revealed that plant height, canopy spread, primary and secondary branches, chlorophyll content, day to maturity, grain yield and yield components were significantly affected by irrigation regimes. However, there was no statistically significant difference between full irrigation and SI for number of pods per plant, number of seeds per pod, 100-grain weight, grain yield per unit area and grain filling rate. The seed yield of the genotypes when grown under the full irrigation condition increased at a rate of 58% over those in rain-fed condition. Investigation of grain yield and drought resistance indices revealed that FLIP 98-106C and Arman can be selected as the best tolerant genotypes to rain-fed condition. In general, under semi-arid conditions and where some limited water resources are available, SI could be an efficient management practice for alleviating the unfavourable effects of soil moisture stress on the yield of rain-fed chickpea during crucial reproductive growth stages.

scholarly journals Using UAV-based hyperspectral imaging and functional regression to assist in predicting grain yield and related traits in wheat under heat-related stress environments for the purpose of stable yielding genotypes

2021 ◽  
Author(s):  
Lucas Costa ◽  
Jordan McBreen ◽  
Yiannis Ampatzidis ◽  
Jia Guo ◽  
Mostafa Reisi Gahrooei ◽  
...  
Keyword(s):  
Heat Stress ◽  
Grain Yield ◽  
Hyperspectral Imaging ◽  
Complex Traits ◽  
Small Sample ◽  
Data Availability ◽  
Physiological Traits ◽  
Functional Regression ◽  
Yield Prediction ◽  
Genetic Gains

AbstractQuantifying certain physiological traits under heat-stress is crucial for maximizing genetic gain for wheat yield and yield-related components. In-season estimation of different physiological traits related to heat stress tolerance can ensure the finding of germplasm, which could help in making effective genetic gains in yield. However, estimation of those complex traits is time- and labor-intensive. Unmanned aerial vehicle (UAV) based hyperspectral imaging could be a powerful tool to estimate indirectly in-season genetic variation for different complex physiological traits in plant breeding that could improve genetic gains for different important economic traits, like grain yield. This study aims to predict in-season genetic variations for cellular membrane thermostability (CMT), yield and yield related traits based on spectral data collected from UAVs; particularly, in cases where there is a small sample size to collect data from and a large range of features collected per sample. In these cases, traditional methods of yield-prediction modeling become less robust. To handle this, a functional regression approach was employed that addresses limitations of previous techniques to create a model for predicting CMT, grain yield and other traits in wheat under heat stress environmental conditions and when data availability is constrained. The results preliminarily indicate that the overall models of each trait studied presented a good accuracy compared to their data’s standard deviation. The yield prediction model presented an average error of 13.42%, showing the function-on-function algorithm chosen for the model as reliable for small datasets with high dimensionality.

Sign in / Sign up

Export Citation Format

Share Document