Soybean Yield Prediction Research Articles

A Hybrid Model for Soybean Yield Prediction Integrating Convolutional Neural Networks, Recurrent Neural Networks, and Graph Convolutional Networks

Soybean yield prediction is one of the most critical activities for increasing agricultural productivity and ensuring food security. Traditional models often underestimate yields because of limitations associated with single data sources and simplistic model architectures. These prevent complex, multifaceted factors influencing crop growth and yield from being captured. In this line, this work fuses multi-source data—satellite imagery, weather data, and soil properties—through the approach of multi-modal fusion using Convolutional Neural Networks and Recurrent Neural Networks. While satellite imagery provides information on spatial data regarding crop health, weather data provides temporal insights, and the soil properties provide important fertility information. Fusing these heterogeneous data sources embeds an overall understanding of yield-determining factors in the model, decreasing the RMSE by 15% and improving R2 by 20% over single-source models. We further push the frontier of feature engineering by using Temporal Convolutional Networks (TCNs) and Graph Convolutional Networks (GCNs) to capture time series trends, geographic and topological information, and pest/disease incidence. TCNs can capture long-range temporal dependencies well, while the GCN model has complex spatial relationships and enhanced the features for making yield predictions. This increases the prediction accuracy by 10% and boosts the F1 score for low-yield area identification by 5%. Additionally, we introduce other improved model architectures: a custom UNet with attention mechanisms, Heterogeneous Graph Neural Networks (HGNNs), and Variational Auto-encoders. The attention mechanism enables more effective spatial feature encoding by focusing on critical image regions, while the HGNN captures interaction patterns that are complex between diverse data types. Finally, VAEs can generate robust feature representation. Such state-of-the-art architectures could then achieve an MAE improvement of 12%, while R2 for yield prediction improves by 25%. In this paper, the state of the art in yield prediction has been advanced due to the employment of multi-source data fusion, sophisticated feature engineering, and advanced neural network architectures. This provides a more accurate and reliable soybean yield forecast. Thus, the fusion of Convolutional Neural Networks with Recurrent Neural Networks and Graph Networks enhances the efficiency of the detection process.

Integration of Optical and Synthetic Aperture Radar Data with Different Synthetic Aperture Radar Image Processing Techniques and Development Stages to Improve Soybean Yield Prediction

Predicting crop yield throughout its development cycle is crucial for planning storage, processing, and distribution. Optical remote sensing has been used for yield prediction but has limitations, such as cloud interference and only capturing canopy-level data. Synthetic Aperture Radar (SAR) complements optical data by capturing information even in cloudy conditions and providing additional plant insights. This study aimed to explore the correlation of SAR variables with soybean yield at different crop stages, testing if SAR data enhances predictions compared to optical data alone. Data from three growing seasons were collected from an area of 106 hectares, using eight SAR variables (Alpha, Entropy, DPSVI, RFDI, Pol, RVI, VH, and VV) and four speckle noise filters. The Random Forest algorithm was applied, combining SAR variables with the EVI optical index. Although none of the SAR variables showed strong correlations with yield (r < |0.35|), predictions improved when SAR data were included. The best performance was achieved using DPSVI with the Boxcar filter, combined with EVI during the maturation stage (with EVI:RMSE = 0.43, 0.49, and 0.60, respectively, for each season; while EVI + DPSVI:RMSE = 0.39, 0.49, and 0.42). Despite improving predictions, the computational demands of SAR processing must be considered, especially when optical data are limited due to cloud cover.

Early Modeling of the Upcoming Landsat Next Constellation for Soybean Yield Prediction Under Varying Levels of Water Availability

The upcoming Landsat Next will provide more frequent land surface observations at higher spatial and spectral resolutions that will greatly benefit the agricultural sector. Early modeling of the upcoming Landsat Next products for soybean yield prediction is essential for long-term satellite monitoring strategies. In this context, this article evaluates the contribution of Landsat Next’s improved spectral resolution for soybean yield prediction under varying levels of water availability. Ground-based hyperspectral data collected over five cropping seasons at the Brazilian Agricultural Research Corporation were resampled to Landsat Next spectral resolution. The spectral dataset (n = 384) was divided into calibration and external validation datasets and investigated using three strategies for soybean yield prediction: (1) using the reflectance from each spectral band; (2) using existing and new vegetation indices developed based on three general equations: Normalized Difference Vegetation Index (NDVI-like), Band Ratio Vegetation Index (RVI-like), and Band Difference Vegetation Index (DVI-like), replacing the traditional spectral bands by all possible combinations between two bands for index calculation; and (3) using a partial least squares regression (PLSR) model composed of all Landsat Next spectral bands, in comparison to PLSR models using Landsat OLI and Sentienel-2 MSI bands. The results show the distribution of the new spectral bands over the most prominent changes in leaf reflectance due to water deficit, particularly in the visible and shortwave infrared spectrum. (1) Band 18 (centered at 1610 nm) had the highest correlation with yield (R2 = 0.34). (2) A new vegetation index, called Normalized Difference Shortwave Vegetation Index (NDSWVI), is proposed and calculated from bands 19 and 20 (centered at 2028 and 2108 nm). NDSWVI showed the best performance (R2 = 0.37) compared to traditional existing and new vegetation indices. (3) The PLSR model gave the best results (R2 = 0.65), outperforming the Landsat OLI and Sentinel-2 MSI sensors. The improved spectral resolution of Landsat Next is expected to contribute to improved crop monitoring, especially for soybean crops in Brazil, increasing the sustainability of the production systems and strengthening food security in Brazil and globally.

Development of an Ensemble Learning Approach for Soybean Yield Prediction using Satellite and Meteorological Data

Development of an Ensemble Learning Approach for Soybean Yield Prediction using Satellite and Meteorological Data

Using gross primary production data and deep transfer learning for crop yield prediction in the US Corn Belt

Estimating crop yield prior to harvest plays a crucial role in agricultural decision making. Machine learning models that use remote sensing (RS) data can provide quick and early estimates of crop yield. Previously, timeseries gross primary production (GPP) synthesized from RS data has been employed in agricultural applications. However, relevant research on using GPP data and deep learning models in crop yield prediction is scarce. The aim of this study was to assess the capability of MODIS-derived GPP data and deep transfer learning to predict crop yield. We developed and trained deep neural network (DNN), one-dimensional convolutional neural network (1D-CNN), and gated recurrent unit (GRU) models to predict county-level corn and soybean yields. Then we evaluated the performance of crop-type and location transfer learning strategies in crop yield prediction. The results of this study can be summarized as follows: (1) GPP data and deep transfer learning have the potential to predict crop yield with reasonable prediction accuracy; (2) in the context of transfer learning, crop-type transfer learning produced a coefficient of determination (R2) ranging from 0.521 to 0.784 for soybean and from 0.644 to 0.903 for corn. Meanwhile, location transfer learning yielded R2 values ranging from 0.690 to 0.931 for the Great Plains (GP) climatic zone and 0.480 to 0.801 for the Eastern Temperate Forest (ETF) climatic zone; and (3) For corn and soybean yield predictions, the 1D-CNN model (R2 = 0.864 for corn and 0.750 for soybean) outperformed the DNN and GRU models. Our results reveal that MODIS-derived GPP data and deep transfer learning can be used to effectively predict county-level crop yield.

Estimation of soybean yield based on high-throughput phenotyping and machine learning.

Soybeans are an important crop used for food, oil, and feed. However, China's soybean self-sufficiency is highly inadequate, with an annual import volume exceeding 80%. RGB cameras serve as powerful tools for estimating crop yield, and machine learning is a practical method based on various features, providing improved yield predictions. However, selecting different input parameters and models, specifically optimal features and model effects, significantly influences soybean yield prediction. This study used an RGB camera to capture soybean canopy images from both the side and top perspectives during the R6 stage (pod filling stage) for 240 soybean varieties (a natural population formed by four provinces in China: Sichuan, Yunnan, Chongqing, and Guizhou). From these images, the morphological, color, and textural features of the soybeans were extracted. Subsequently, feature selection was performed on the image parameters using a Pearson correlation coefficient threshold ≥0.5. Five machine learning methods, namely, CatBoost, LightGBM, RF, GBDT, and MLP, were employed to establish soybean yield estimation models based on the individual and combined image parameters from the two perspectives extracted from RGB images. (1) GBDT is the optimal model for predicting soybean yield, with a test set R2 value of 0.82, an RMSE of 1.99 g/plant, and an MAE of 3.12%. (2) The fusion of multiangle and multitype indicators is conducive to improving soybean yield prediction accuracy. Therefore, this combination of parameters extracted from RGB images via machine learning has great potential for estimating soybean yield, providing a theoretical basis and technical support for accelerating the soybean breeding process.

Global de-trending significantly improves the accuracy of XGBoost-based county-level maize and soybean yield prediction in the Midwestern United States

ABSTRACT The application of machine learning in crop yield prediction has gained considerable traction, yet uncertainties persist regarding the impact of the yield trends on these predictions and the differences between the detrending methods. In our study, we utilized extreme gradient boosting (XGBoost) to scrutinize the effects of no trend processing (NTP), input year as a feature (IYF), input average yield as a feature (IAYF), input linear yield as a feature (ILYF), and the global detrending method (GDT) on the yield prediction of maize and soybean in the Midwestern United States. Based on our findings, compared with that of NTP, the incorporation of the yield trend as a predictor in XGBoost significantly improved the accuracy and reduced the uncertainty of the yield prediction. Notably, GDT emerged as a standout performer, significantly reducing the average yield prediction error by 0.091 t/ha for soybean and 0.158 t/ha for maize with respect to NTP, and concurrently improving the determination coefficient (R2) by 20.6% and 19.6% for soybean and maize, respectively. Compared with IYF, IAYF, and ILYF, GDT showed substantial improvements ranging from 3.8% to 12.7% in R2 for soybean and 3.6% to 12.7% for maize. The SHapley Additive ExPlanations (SHAP) framework showed that the enhanced vegetation index (EVI), particularly during the soybean podding and maize dough formation stages, played a crucial role in understanding the variations in interannual yield variability. These findings confirmed the importance of GDT in crop yield prediction via machine learning and could be used to facilitate future advancements in machine learning applications for yield forecasting.

Satellite-based soybean yield prediction in Argentina: A comparison between panel regression and deep learning methods

The accurate prediction of soybean yield is vital for global food market stabilization and food security. Recent advancements in remote sensing technology have significantly amplified interest in leveraging satellite-based methods for predicting crop yield. These methods offer in-season yield estimates. By utilizing this timely information, decision-makers can formulate strategic, well-informed choices that preemptively mitigate potential food price hikes, ultimately bolstering food security. While simple regression models have been widely utilized for satellite-based yield prediction, researchers have recently begun to explore the use of deep learning algorithms. This study compares the performance of panel regression and deep learning models for in-season soybean yield prediction at the Department (county-equivalent) level in Argentina. Data sources include the latest soybean land use products and MODIS bi-weekly vegetation index products. Results indicate that deep learning models significantly outperform panel regression. Deep learning Long Short-Term Memory (LSTM) models, which incorporate attention mechanism and a series of peak NDVI images, generate more accurate and time-sensitive predictions. Among competing LSTM models, the one with attention mechanism applied to the entire growing season's NDVI data yields the highest prediction accuracy, with a Root Mean Square Error (RMSE) of 505.78 kg/ha and Normalized Root Mean Square Error (NRMSE) of 0.0726. The LSTM model with attention on the three highest NDVI images attains a satisfactory prediction accuracy (RMSE = 627.28 kg/ha, NRMSE = 0.089) six weeks prior to harvest. This study presents a robust model for predicting crop yields, promoting sustainable production of soybeans and facilitating knowledgeable choices among farmers and policymakers.

A Bayesian Regression Approach for Estimating Photosynthetically Active Radiation Using Satellite Data: Implications for Soybean Yield Prediction using the CROPGRO Model

A Bayesian Regression Approach for Estimating Photosynthetically Active Radiation Using Satellite Data: Implications for Soybean Yield Prediction using the CROPGRO Model

Beyond assimilation of leaf area index: Leveraging additional spectral information using machine learning for site-specific soybean yield prediction

Assimilating external observations of crop state in cropping system models is essential for making spatially explicit predictions of crop variables relevant in precision agriculture. Satellite-based leaf area index (LAI) estimates have been the most frequent variable used as a proxy of actual crop growth. However, additional information beyond LAI, like canopy N content, water content, and structure, can be retrieved from satellite observations. Including such variables by data assimilation directly is difficult because many crop models do not have corresponding state variables or the relationship between the observations and the process that regulates crop growth is unclear. Therefore, other approaches are required to include such information. In this study, we investigate the improvement in the predicted yield and feature impact on model outputs by using a hybrid approach that combines observations from Sentinel-1 and 2 time-series with the outputs from a process-based model embedded in a data assimilation framework and uses the Gradient-boosted trees regressor (GBTR) as predictive model. We used two regions with soybean fields: the US (13 K points) and Uruguay (400 K points). We found an advantage when using the GBTR as the predictive model (reduced RRMSE by ∼16%) compared to data assimilation. Adding the vegetation indices had a marginal improvement (reduced RRMSE by ∼1%), while the impact of adding reflectance and backscatter values was negative. The satellite-based features had a very small importance score, while features' impact on prediction was predominantly unclear, explaining the marginal predictive power added by satellite-based features. We found that features from the reproductive stages had the highest importance, while the importance of an index related to drought stress (NMDI) across the growing season provided insights for further improvement of data assimilation methods. However, more studies are required to better disentangle pathways towards further improvement in constraining crop models by ingesting satellite observations.

Bridging the gap between crop breeding and GeoAI: Soybean yield prediction from multispectral UAV images with transfer learning

Despite significant progress has been made towards crop yield prediction with remote sensing, there exist knowledge gaps on (1) the impacts of temporal resolution of imaging frequencies on yield prediction, (2) transferability of the models among different genotypes and test sites, and (3) translation of these research developments to crop breeding that benefit farmers. Existing research predominantly provides an on-site perspective, frequently missing the complexities of real-world applications. The objectives of this paper are to investigate the transferability and generalization capabilities of yield prediction models for crop breeding across test sites located in North and South Americas. Toward that goal, we tested different machine learning techniques including Random Forest Regressor (RF), Gradient Boosting Regression (GB), and Deep Neural Networks (DNN) for soybean yield prediction with experiments conducted in different climate and growth conditions. A novel transfer learning approach was proposed for genotype selection and categorizing soybean yield for screening high-yield varieties. Furthermore, we studied the effect of temporal resolution on yield prediction, focusing on the critical development stages and optimal aerial survey frequencies for precise yield prediction using large 31,404 sample data. Results demonstrated that the combined dataset of Argentina and United States representing different climate regimes provided the highest performance with an R2 of 0.76 using RF and GB algorithms. The classification approach was proven to be most useful for crop breeding as demonstrated by accurately identifying the high-yielding genotypes. Increasing temporal sampling of key phenological stages significantly improved yield prediction. Although transfer learning yielded promising outcomes across trials within Argentina the efficacy of transferring models from Argentina to the United States was limited, attributed to significant seasonal and climate variations. This study pioneered the use of transfer learning for model adaptability in real-world breeding scenarios, training and transferring models within the South and North Americas, providing actionable insights and strategies for the breeding community, aiming to facilitate improved decision-making for agricultural productivity.

Application of UAV Multispectral Imaging to Monitor Soybean Growth with Yield Prediction through Machine Learning

The application of remote sensing, which is non-destructive and cost-efficient, has been widely used in crop monitoring and management. This study used a built-in multispectral imager on a small unmanned aerial vehicle (UAV) to capture multispectral images in five different spectral bands (blue, green, red, red edge, and near-infrared), instead of satellite-captured data, to monitor soybean growth in a field. The field experiment was conducted in a soybean field at the Mississippi State University Experiment Station near Pontotoc, MS, USA. The experiment consisted of five cover crops (Cereal Rye, Vetch, Wheat, Mustard plus Cereal Rye, and native vegetation) planted in the winter and three fertilizer treatments (Fertilizer, Poultry Liter, and None) applied before planting the soybean. During the soybean growing season in 2022, eight UAV imaging flyovers were conducted, spread across the growth season. UAV image-derived vegetation indices (VIs) coupled with machine learning (ML) models were computed for characterizing soybean growth at different stages across the season. The aim of this study focuses on monitoring soybean growth to predict yield, using 14 VIs including CC (Canopy Cover), NDVI (Normalized Difference Vegetation Index), GNDVI (Green Normalized Difference Vegetation Index), EVI2 (Enhanced Vegetation Index 2), and others. Different machine learning algorithms including Linear Regression (LR), Support Vector Machine (SVM), and Random Forest (RF) are used for this purpose. The stage of the initial pod development was shown as having the best predictability for earliest soybean yield prediction. CC, NDVI, and NAVI (Normalized area vegetation index) were shown as the best VIs for yield prediction. The RMSE was found to be about 134.5 to 511.11 kg ha−1 in the different yield models, whereas it was 605.26 to 685.96 kg ha−1 in the cross-validated models. Due to the limited number of training and testing samples in the K-fold cross-validation, the models’ results changed to some extent. Nevertheless, the results of this study will be useful for the application of UAV remote sensing to provide information for soybean production and management. This study demonstrates that VIs coupled with ML models can be used in multistage soybean yield prediction at a farm scale, even with a limited number of training samples.

Critical evaluation of the effects of a cross-validation strategy and machine learning optimization on the prediction accuracy and transferability of a soybean yield prediction model using UAV-based remote sensing

Crop yield prediction models are critical tools for evaluating growth performance and informing decisions during farm management. Developing yield prediction models that are robust not only in the ranges of the model spatial domain but also in additional locations using a data-driven approach is challenging. The main objective of this study was to investigate an appropriate cross-validation (CV) strategy for establishing transferable UAV-based yield prediction models across different spatial domains (i.e., meeting extrapolation mapping objectives). In this study, we compared three data splitting procedures for the CV protocols, including random data splitting (random CV), cluster-based spatial splitting (spatial CV), and field-specific hold-out data splitting (leave-one-field-out CV). Model optimization was also examined to determine whether these factors affect the transferability of the yield model, including performing recursive feature elimination (RFE) and comparing the effects of algorithms utilized in the yield prediction model. Three base learner algorithms, namely, random forest, XGBoost, and LASSO regression, were utilized, and a stacked ensemble technique model formed with these base learners was also implemented. The established models were later tested on an independent field as a test dataset to evaluate the model transferability performance. Random CV exhibited poor error tracking performance in predicting yield beyond the model spatial domain, while spatial CV and leave-one-field-out CV approaches provided better expectation on yield predictions outside the model's training spatial domain. Furthermore, simple models as implementing LASSO regression and RFE improved the model capability in extrapolation tasks. The results of this study suggest that spatially-aware CV should be used as the standard method rather than conventional random CV for validating the yield model to ensure a more realistic and reliable yield model in extrapolation objectives.

Improving soybean yield prediction by integrating UAV nadir and cross-circling oblique imaging

High-throughput estimation of soybean yield using unmanned aerial vehicle (UAV) imagery can help improve the efficiency of soybean breeding. Previous studies have mainly focused on the extraction of vegetation indices and texture features from two-dimensional(2D) orthophotos to construct empirical models of yield, lacking spatial structure information of crops. Therefore, UAV cross-circling oblique (CCO) photography combined with SfM-MVS algorithm was used to reconstruct three-dimensional(3D) soybean canopy structure. Then canopy 3D related phenotypic features are extracted and combined with features from RGB nadir and multispectral images to analyze the capability of different modal data fusion on soybean yield prediction. In addition, Shapley value was used to evaluate the importance of features across different machine learning models. Based on the Shapley value, a bagging-stacking ensemble learning framework was developed using Lasso, Random Forest (RF), Ridge Regression (RR), and XGBoost as base learners for yield prediction. The performance of the traditional stacking method was evaluated and compared with weighted average methods such as Bayesian Model Averaging (BMA) and Entropy Weighted Average (EWA) as meta-learners in the ensemble framework. The results show that 3D canopy structure of soybean can be obtained from UAV CCO photography. The inclusion of 3D structural features can improve the accuracy of yield estimation. Among different modal data combinations, the highest estimation accuracy was achieved when combining RGB nadir features with CCO 3D features. The performance of the above base learners was improved by 8.8%, 3.5%, 7.1%, and 8.0% respectively when using the Shapley value method. The accuracy of yield prediction applied on independent dataset of year 2023 was further calculated by using the bagging-stacking ensemble learning method. When using BMA as the meta-learner, the best performance is obtained with an R2 of 0.7. Therefore, UAV CCO photography with SfM-MVS algorithm provides a new approach to obtain high-quality point clouds of the crop canopy at low cost. UAV multimodal data combined with ensemble learning models allow accurate estimation of yield prediction in breeding materials of soybeans at the plot scale.

Leveraging Remote Sensing Data for Yield Prediction with Deep Transfer Learning

Remote sensing data represent one of the most important sources for automized yield prediction. High temporal and spatial resolution, historical record availability, reliability, and low cost are key factors in predicting yields around the world. Yield prediction as a machine learning task is challenging, as reliable ground truth data are difficult to obtain, especially since new data points can only be acquired once a year during harvest. Factors that influence annual yields are plentiful, and data acquisition can be expensive, as crop-related data often need to be captured by experts or specialized sensors. A solution to both problems can be provided by deep transfer learning based on remote sensing data. Satellite images are free of charge, and transfer learning allows recognition of yield-related patterns within countries where data are plentiful and transfers the knowledge to other domains, thus limiting the number of ground truth observations needed. Within this study, we examine the use of transfer learning for yield prediction, where the data preprocessing towards histograms is unique. We present a deep transfer learning framework for yield prediction and demonstrate its successful application to transfer knowledge gained from US soybean yield prediction to soybean yield prediction within Argentina. We perform a temporal alignment of the two domains and improve transfer learning by applying several transfer learning techniques, such as L2-SP, BSS, and layer freezing, to overcome catastrophic forgetting and negative transfer problems. Lastly, we exploit spatio-temporal patterns within the data by applying a Gaussian process. We are able to improve the performance of soybean yield prediction in Argentina by a total of 19% in terms of RMSE and 39% in terms of R2 compared to predictions without transfer learning and Gaussian processes. This proof of concept for advanced transfer learning techniques for yield prediction and remote sensing data in the form of histograms can enable successful yield prediction, especially in emerging and developing countries, where reliable data are usually limited.

End-to-end 3D CNN for plot-scale soybean yield prediction using multitemporal UAV-based RGB images

Crop yield prediction from UAV images has significant potential in accelerating and revolutionizing crop breeding pipelines. Although convolutional neural networks (CNN) provide easy, accurate and efficient solutions over traditional machine learning models in computer vision applications, a CNN training requires large number of ground truth data, which is often difficult to collect in the agricultural context. The major objective of this study was to develope an end-to-end 3D CNN model for plot-scale soybean yield prediction using multitemporal UAV-based RGB images with approximately 30,000 sample plots. A low-cost UAV-RGB system was utilized and multitemporal images from 13 different experimental fields were collected at Argentina in 2021. Three commonly used 2D CNN architectures (i.e., VGG, ResNet and DenseNet) were transformed into 3D variants to incorporate the temporal data as the third dimension. Additionally, multiple spatiotemporal resolutions were considered as data input and the CNN architectures were trained with different combinations of input shapes. The results reveal that: (a) DenseNet provided the most efficient result (R2 0.69) in terms of accuracy and model complexity, followed by VGG (R2 0.70) and ResNet (R2 0.65); (b) Finer spatiotemporal resolution did not necessarily improve the model performance but increased the model complexity, while the coarser resolution achieved comparable results; and (c) DenseNet showed lower clustering patterns in its prediction maps compared to the other models. This study clearly identifies that multitemporal observation with UAV-based RGB images provides enough information for the 3D CNN architectures to accurately estimate soybean yield non-destructively and efficiently.

3D Convolutional Neural Network with Dimension Reduction and Metric Learning for Crop Yield Prediction Based on Remote Sensing Data

Crop yield prediction is essential for tasks like determining the optimal profile of crops to be planted, allocating government resources, effectively planning and preparing for aid distribution, making decisions about imports, and so on. Crop yield prediction using remote sensing data during the growing season is helpful to farm planning and management, which has received increasing attention. Information mining from multichannel geo-spatiotemporal data brings many benefits to crop yield prediction. However, most of the existing methods have not fully utilized the dimension reduction technology and the spatiotemporal feature of the data. In this paper, a new approach is proposed to predict the yield from multispatial images by using the dimension reduction method and a 3D convolutional neural network. In addition, regions with similar crop yields should have similar features learned by the network. Thus, metric learning and multitask learning are used to learn more discriminative features. We evaluated the proposed method on county-level soybean yield prediction in the United States, and the experimental results show the effectiveness of the proposed method. The proposed method provides new ideas for crop yield estimation and effectively improves the accuracy of crop yield estimation.

3D-ResNet-BiLSTM Model: A Deep Learning Model for County-Level Soybean Yield Prediction with Time-Series Sentinel-1, Sentinel-2 Imagery, and Daymet Data

Ensuring food security in precision agriculture requires early prediction of soybean yield at various scales within the United States (U.S.), ranging from international to local levels. Accurate yield estimation is essential in preventing famine by providing insights into food availability during the growth season. Numerous deep learning (DL) algorithms have been developed to estimate soybean yield effectively using time-series remote sensing (RS) data to achieve this goal. However, the training data with short time spans can limit their ability to adapt to the dynamic and nuanced temporal changes in crop conditions. To address this challenge, we designed a 3D-ResNet-BiLSTM model to efficiently predict soybean yield at the county level across the U.S., even when using training data with shorter periods. We leveraged detailed Sentinel-2 imagery and Sentinel-1 SAR images to extract spectral bands, key vegetation indices (VIs), and VV and VH polarizations. Additionally, Daymet data was incorporated via Google Earth Engine (GEE) to enhance the model’s input features. To process these inputs effectively, a dedicated 3D-ResNet architecture was designed to extract high-level features. These enriched features were then fed into a BiLSTM layer, enabling accurate prediction of soybean yield. To evaluate the efficacy of our model, its performance was compared with that of well-known models, including the Linear Regression (LR), Random Forest (RF), and 1D/2D/3D-ResNet models, as well as a 2D-CNN-LSTM model. The data from a short period (2019 to 2020) were used to train all models, while their accuracy was assessed using data from the year 2021. The experimental results showed that the proposed 3D-Resnet-BiLSTM model had a superior performance compared to the other models, achieving remarkable metrics (R2 = 0.791, RMSE = 5.56 Bu Ac−1, MAE = 4.35 Bu Ac−1, MAPE = 9%, and RRMSE = 10.49%). Furthermore, the 3D-ResNet-BiLSTM model showed a 7% higher R2 than the ResNet and RF models and an enhancement of 27% and 17% against the LR and 2D-CNN-LSTM models, respectively. The results highlighted our model’s potential for accurate soybean yield predictions, supporting sustainable agriculture and food security.

Artificial intelligence and satellite‐based remote sensing can be used to predict soybean (Glycine max) yield

AbstractBecause the manual counting of soybean (Glycine max) plants, pods, and seeds/pods is unsuitable for soybean yield predictions, alternative methods are desired. Therefore, the objective was to determine if satellite remote sensing‐based artificial intelligence (AI) models could be used to predict soybean yield. In the study, multiple remote sensing‐based AI models were developed for soybean growth stage ranging from VE/VC (plant emergence) to R6/R7 (full seed to beginning maturity). The ability of the deep neural network (DNN), support vector machine (SVM), random forest (RF), least absolute shrinkage and selection operator (LASSO), and AdaBoost to predict soybean yield, based on blue, green, red, and near‐infrared reflectance data collected by the PlanetScope satellite at six growth stages, was determined. Remote sensing and soybean yield monitor data from three different fields in 2 years (2019 and 2021) were aggregated into 24,282 grid cells that had the dimensions of 10 m by 10 m. A comparison across models showed that the DNN outperformed the other models. Moreover, as crops matured from VE/VC to R4/R5, the R2 value of the models increased from 0.26 to over 0.70. These findings indicate that remote sensing data collected at different growth stages can be combined for soybean yield predictions. Moreover, additional work needs to be conducted to assess the model's ability to predict soybean yield with vegetation indices (VIs) data for fields not used to train the model.

Estimation of Soybean Yield by Combining Maturity Group Information and Unmanned Aerial Vehicle Multi-Sensor Data Using Machine Learning

Accurate and rapid estimation of the crop yield is essential to precision agriculture. Critical to crop improvement, yield is a primary index for selecting excellent genotypes in crop breeding. Recently developed unmanned aerial vehicle (UAV) platforms and advanced algorithms can provide powerful tools for plant breeders. Genotype category information such as the maturity group information (M) can significantly influence soybean yield estimation using remote sensing data. The objective of this study was to improve soybean yield prediction by combining M with UAV-based multi-sensor data using machine learning methods. We investigated three types of maturity groups (Early, Median and Late) of soybean, and collected the UAV-based hyperspectral and red–green–blue (RGB) images at three key growth stages. Vegetation indices (VI) and texture features (Te) were extracted and combined with M to predict yield using partial least square regression (PLSR), Gaussian process regression (GPR), random forest regression (RFR) and kernel ridge regression (KRR). The results showed that (1) the method of combining M with remote sensing data could significantly improve the estimation performances of soybean yield. (2) The combinations of three variables (VI, Te and M) gave the best estimation accuracy. Meanwhile, the flowering stage was the optimal single time point for yield estimation (R2 = 0.689, RMSE = 408.099 kg/hm2), while using multiple growth stages produced the best estimation performance (R2 = 0.700, RMSE = 400.946 kg/hm2). (3) By comparing the models constructed by different algorithms for different growth stages, it showed that the models built by GPR showed the best performances. Overall, the results of this study provide insights into soybean yield estimation based on UAV remote sensing data and maturity information.