Rice Yield Prediction Research Articles

Artificial Bee Colony Algorithm-based Feature Selection and Hybrid ML Framework for Efficient Rice Yield Prediction

India's economy predominantly depends on monsoon and agricultural output. Agribusiness products contribute to nearly a quarter of its gross domestic product and 58% of its population depends on agriculture for their livelihood. Certain crops, like rice, are vital to its food security being the most widely grown crop and accounting for one-third production of foodgrains in India. Understanding and enhancing its production is critical in ensuring food availability and promoting sustainable agricultural practices. Rice yield prediction has been a most researched area in the agriculture domain. Machine Learning (ML) frameworks have been found to perform well in patches with large, complex datasets as insufficient feature engineering and temporal dependencies plague efficacy. In this paper, we propose a swam-based meta-heuristic artificial bee colony (ABC) algorithm for feature selection from the dataset sourced from the Agricultural Production and Statistical Division of the Department of Agriculture Cooperation and Farmers Welfare, Government of India. The feature engineering is further optimized by a hybrid model comprising a convolutional neural network (CNN) for learning hierarchical representations and identifying relevant attributes from the complex dataset and long short-term memory (LSTM) for temporal aspects. Finally, a random forest (RF) regressor provides the benefits of ensemble learning, which merges multiple decision trees to remove bias, and variance and improve prediction accuracy. From the results, it is observed that the proposed hybrid model outperforms existing state-of-the-art standalone and hybrid models with the highest coefficient of determination (R2) and lowest mean square error (MSE) of 0.989 and 13613 respectively. The reliable and efficient hybrid model can aid farmers and policymakers in making informed decisions related to rice yield prediction leading to sustainable agricultural practices.

Zone-II & III: Machine Learning based Rice Yield Prediction in Andhra Pradesh

In a unique approach, this research predicts rice yield in Zones II and III of Andhra Pradesh using machine learning algorithms. Because food security and agricultural sustainability are becoming more and more important, an accurate assessment of crop yields is necessary for agricultural planning and decision-making. By using machine learning approaches, we hope to develop robust prediction models that can forecast rice yields depending on a variety of agronomic and environmental conditions. Our plan includes collecting a lot of data on variables such as soil qualities, weather patterns, crop management practices, and historical yield records from the target regions. Using careful feature engineering and data preprocessing, we find relevant predictors and derive significant insights for rice yield estimation. Next, we employ state-of-the-art machine learning methods, like Support Vector Machines, Random Forest, and Gradient Boosting, to build trustworthy prediction models. These models are trained on historical data and then cross-validated to guarantee accuracy and generalizability. Our goal is to combine domain expertise in agriculture with cutting-edge machine learning techniques to provide farmers, policymakers, and other agricultural stakeholders with insightful understandings of the dynamics of rice production. In Zones II and III of Andhra Pradesh, this will make it easier to allocate resources and make improve agricultural output and well-informed decisions sustainability.

Comparative analysis of SMLR, ANN, Elastic net and LASSO based models for rice crop yield prediction in Uttarakhand

The study was aimed to develop the yield forecast model for rice crop yield. Four different techniques i.e. Stepwise Multiple Linear Regression (SMLR), Artificial Neural Network (ANN), Least Absolute Shrinkage and Selection Operator (LASSO) and Elastic Net (ELNET)were used to build the prediction models. Dataset of meteorological data and crop yield data of 15 years have been used to develop the forecast models. The developed models were also validated on the dataset of three years. The assessment of the developed models wasdone by using root mean square error (RMSE),normalized root mean square error (nRMSE),Mean Absolute Error (MAE) and on the basis of coefficient of determination (R2). The experimental analysis suggested that the performance for Artificial Neural Network (R2=0.99, RMSE=0.07, nRMSE=2.20, MAE=0.06) is better as compared to SMLR(R2=0.97, RMSE=0.08, nRMSE=2.34, MAE=0.05), LASSO (R2=0.62, RMSE=0.26, nRMSE=7.81, MAE=0.24) and ELNET (R2=0.54, RMSE=0.38, nRMSE=11.41, MAE=0.37) for the predictionof rice crop yield for Udham Singh Nagar (USN) district of Uttarakhand. Therefore, for the prediction of rice yield, ANN technique can be well utilised for Udham Singh Nagar district of Uttarakhand.

Enhancing direct-seeded rice yield prediction using UAV-derived features acquired during the reproductive phase

Enhancing direct-seeded rice yield prediction using UAV-derived features acquired during the reproductive phase

Future climate change may pose pressures on greenhouse gas emission reduction in China’s rice production

Paddy rice cultivation is a vital food source but also a significant contributor to greenhouse gas (GHG) emissions, particularly methane. Management measures such as improved water management and fertilization are being pursued to sustain rice production while cutting down the emissions. However, climate change may pose further challenges for sustainable rice agriculture by either impacting rice production or GHG emissions. Based on a database consisting of 1216 observations across China, we established machine learning models predicting rice yield and GHG emissions, taking into account climate, soil, and management variables. We tested three machine learning methods including random forest, support vector machine and artificial neural networks. The models were trained on 70% of the dataset with the remaining 30% used to evaluate the model performance. Random forest performed best with R2 of 0.59 to 0.72 and modelling efficiency of 52% to 72%. Farmland management emerged as the most important variable to the model prediction, followed by climate and soil variables. Rice production and GHG emissions were estimated to be 245.36 Tg and 218.95 Tg CO2-eq in 2018, respectively. By the year 2050, rice yield was expected to experience a modest decrease ranging from 0.6% to 1.2% due to the combined effects of global change variables, but GHG emissions were projected to increase by 2.1% to 4.9% under different climate change scenarios (Representative Concentration Pathways) in China. A large spatial variability in the impact of climate change was observed, and climate change will have the most significant impact in the Northeast agricultural region of China due to the warming effect. Through the analysis of region-specific farmland management scenarios, this study underscores a “Code Red” situation, signifying that future climate change could pose unprecedented risks to the sustainability of rice production in China, despite ongoing management improvements.

Delineation of Rice Yield and Production in Combination of Crop Model and Remote Sensing

The study was to simulate rice yield in various places and outline a rice yield map for the study area using GIS, remote sensing, and a rice model. As a case study, the data were collected on the climate, soil characteristics, and rice cropping status in the AnGiang province (Southern of the Mekong River Delta in Viet Nam). The AquaCrop model was used to predict rice yield. The MODIS image delineated the rice cropping status based on spatial and temporal NDVI values. The results of the yield simulation are then put together with information about where the rice was planted, the weather, and the properties of the soil to make a map of the yield distribution. Finally, the outcomes are verified and contrasted with the statistical findings in the last step. The rice yield was predicted and compared with actual 1 and 6 percent rice yields. The anticipated rice yield map was established for the Winter-Spring cropping season 2012-2013 and the Summer-Autumn and Autumn–Winter cropping seasons 2013. Rice production and yield distribution can be divided into two major areas. The alluvial soil area produces significantly more rice than the LongXuyen quadrangle area because of the difference in soil and weather conditions. Rice yield simulation and delineation combining remote sensing and crop models is a good approach for yield prediction and better agricultural management strategy development in a country or region. The accuracy of the results depends on the quality of the input data, such as soil weather and remote sensing.
 Keywords: GIS, MODIS, cropping season, rice yield

Improved information dissemination services for the agricultural sector in Thailand: development and evaluation of a machine learning based rice crop yield prediction system

The upward surge in world population, sustainability, and concerns about food and nutritional security place the food and agricultural sector amidst challenges. These challenges encompass increasing food supply, meeting higher quality standards, and enhancing productivity and crop yield prediction. This study discusses the development and evaluation of a Machine Learning (ML) based rice yield prediction system. Utilizing extensive historical datasets from the Office of Agricultural Economics (OAE), Ministry of Agriculture and Cooperatives (MOAC), Thai Meteorological Department, and Department of Internal Trade of Thailand, all pivotal variables at the national rice level were considered. Examining ML models like the Generalized Linear Model (GLM), Feed-Forward Neural Network (FFNN), Support Vector Machine (SVM), and Random Forest (RF), the study proposes a Web-based system to disseminate Thailand's rice information and yield predictions, aiding decision-making processes. Beyond evaluating the performance of each prediction model, user satisfaction was scrutinized. Results reveal that the FFNN, a deep neural network, adeptly handles complex nonlinear relationships in high-dimensional datasets. Despite the FFNN's longer training runtime due to Big-O complexity, it exhibits the shortest execution time for predictions. System usability assessment, based on ten standardized questions, indicates participants found the proposed system marginally acceptable, reporting positive user experiences and feeling confident in system use without perceiving it as inconsistent or cumbersome. This study is significant for enhancing agricultural information dissemination services across diverse sectors, benefiting farmers, wholesalers, retailers, and policymakers.

Combining vegetation, color, and texture indices with hyperspectral parameters using machine-learning methods to estimate nitrogen concentration in rice stems and leaves

Context or problemNitrogen is one of the important elements of crops, which plays a decisive role in crop growth and development and the formation of yields. Monitoring of rice organ-scale nitrogen concentration based on the unmanned aerial vehicle (UAV) images is of great significance for rice field management and yield prediction. Objective or research questionPrevious studies have focused on the use of traditional statistical methods and chlorophyll-related vegetation indices to construct plant nitrogen concentration, with models lacking generalizability. MethodsIn this study, rice field trials of two varieties (NJ9108, YD6) and nitrogen fertilizer treatments (N0-N3: 0, 105, 210 and 315 kg/ha) were conducted for 3 years with manual sampling and UAV digital and hyperspectral images during key fertility periods. Based on the data of the whole growth periods and combined with vegetation indices (VIs), color indices (CIs), hyperspectral parameters (HPs), texture indices (TIs) and machine-learning algorithms, monitoring models of nitrogen concentration at the organ scale of rice were constructed and used to estimate the N content of multiple organs (leaf and stem) of rice at different periods. Field experiments were used to collect the multi-organ nitrogen concentration of rice and the remote sensing (RS) data of UAV during the critical growth period of the two years (2021, 2022), and machine-learning algorithms were used to construct the estimation models. ResultsThe results showed that VIs had good correlations with leaf nitrogen concentration (LNC), stem nitrogen concentration (SNC) and plant nitrogen concentration (PNC), with correlation coefficients (r) of 0.86, 0.74 and 0.81, respectively. Machine learning estimation models combining multiple types of RS indices were more accurate than single parameter models constructed by traditional statistical methods, with the LNC optimal model (R2 = 0.8, RMSE = 3.83 mg/g), the SNC optimal model (R2 = 0.7, RMSE = 2.43 mg/g) and the PNC optimal model (R2 = 0.7, RMSE = 3.19 mg/g). Validated using data from 2020, the machine-learning models were far more accurate than traditional methods. ConclusionsThese results show that the use of multi-source remote sensing data based on machine-learning algorithms can effectively estimate the nitrogen concentration of organs in rice. ImplicationsThis study provides an accurate, stable and universal method for estimating rice nitrogen concentration in rice organs, which can be used as a reference for estimating rice nitrogen concentration in large fields using UAV RS technology.

Performance evaluation of the ORYZA_V3 model for rice growth and organ nitrogen content

AbstractThe ORYZA version 3 (ORYZA_V3) model is commonly used for simulating paddy rice (Oryza sativa L.) growth. However, few studies have investigated the performance of the ORYZA_V3 model for predicting in‐season crop growth variables, especially for the nitrogen (N) translocation process among paddy rice organs in the whole growing season, that is, from sowing to harvest stages. This study examines the prediction accuracy of seven paddy rice growth variables based on one cropping season of field experiment. The results show that (1) the ORYZA_V3 model generates accurate prediction of paddy rice leaf area index, above‐ground biomass, and yield, where root mean square errors (RMSEs) are 0.98–1.11 ha/ha, 2360.5–3028.7 kg/ha, and 492.9–515.2 kg/ha, respectively, coefficient of determination (R2) ranges from 0.76 to 1.00, and index of agreement (D‐index) ranges from 0.50 to 0.98. (2) The prediction accuracy of paddy rice leaf and stem N contents is acceptable, where RMSE ranges from 0.0017 to 0.0041 kg/kg, R2 ranges from 0.82 to 0.96, and D‐index ranges from 0.93 to 0.98. (3) The ORYZA_V3 model cannot generate accurate prediction of rice root and panicle N contents, where RMSE ranges from 0.0007 to 0.0022 kg/kg, R2 ranges from 0.01 to 0.20, and D‐index ranges from 0.17 to 0.65. The structural uncertainty in the N translocation module may be a major reason for the unacceptable crop model prediction. This study demonstrated the substantial potential of the ORYZA_V3 model for simulating and understanding paddy rice growth and organ N translocation.

Crop Yield Assessment Using Field-Based Data and Crop Models at the Village Level: A Case Study on a Homogeneous Rice Area in Telangana, India

Crop yield estimation has gained importance due to its vital significance for policymakers and decision-makers in enacting schemes, ensuring food security, and assessing crop insurance losses due to biotic and abiotic stress. This research focused on rice yield estimation at the field level in the Karimnagar district of Telangana during 2021 and 2022 by employing the leaf area index (LAI) as the primary criterion for integrating remote sensing technology and crop simulation models. Using Sentinel-2 satellite data, the rice crop was mapped with the help of ground data and machine learning algorithms, attaining an accuracy of 93.04%. Crop management data for the DSSAT tool were collected during the field visits; the model results revealed a 0.80 correlation between observed and predicted yields. Due to its strong correlation with LAI (0.82), the normalized difference vegetation index (NDVI) was selected as the critical element for integration with the model. A spatial LAI map was generated using the linear equation developed between the NDVI and LAI. The relationship between LAI and yield was used to create a spatial yield map. The study’s findings show that assimilating remote sensing data with crop models enhances the precision of rice yield prediction for insurance companies and policy- and decision-makers.

Improved prediction of rice yield at field and county levels by synergistic use of SAR, optical and meteorological data

Timely and accurate rice yield prediction over large regions is imperative to making informed decisions on precision crop management and ensuring regional food security. Previous studies on crop yield prediction mainly used of either optical or synthetic aperture radar (SAR) data alone, and few of them investigated the synergistic use of these two data sources. It remains unclear whether and how synergistic use of optical and SAR imagery would improve crop yield prediction over a large region. In addition, various machine learning (ML) algorithms have been used in regional yield prediction, but the existing findings are mostly inconsistent about the optimal ML algorithm that prevails others across years and spatial levels. To solve these issues, this study designed a meta-learning ensemble regression (MLER) framework and developed an efficient method of integrating multi-source data for accurate prediction of rice yield at field and county levels over Jiangsu Province of China.The results demonstrated that SAR data from Sentinel-1 could compensate for the lack of optical data from Sentinel-2 for rice yield prediction, which was further improved by adding the meteorological data. The MLER algorithm exhibited the highest field-level prediction accuracy for Xinghua and Suining (R2 = 0.89, RMSE = 0.54 t/ha), as compared with individual base algorithms (RF: R2 = 0.80, RMSE = 0.72 t/ha; XGBoost: R2 = 0.76, RMSE = 0.80 t/ha; SVR: R2 = 0.74, RMSE = 0.83 t/ha) and the long short-term memory (LSTM) (R2 = 0.76, RMSE = 0.80 t/ha) with a ten-fold cross-validation. Furthermore, integrating the MLER algorithm with the best-performing data combination could reduce the influence of regional phenological variability for building yield prediction models applicable to the entire province. The MLER models for predicting the rice yield across the province exhibited robust performance in the leave-one-year-out accuracy assessment at both levels (field level: R2: 0.50∼0.61, RMSE: 0.85∼1.22 t/ha; county level: R2: 0.42∼0.67, RMSE: 0.23∼0.27 t/ha) and could be transferred to an extreme-heat year (2022) (R2 = 0.49, RMSE = 1.52 t/ha). This research opens up new possibilities for predicting the rice yield of smallholder fields over large regions at multiple levels by integrating publicly available multi-source data and machine learning.

Weather-Based Rice Crop Yield Forecasting using Different Regression Techniques & Neural Network Approach for Prayagraj Region

Rice crop yield data and weather data were considered in this study, covering the past twenty-nine years (1991-2019) in Prayagraj District, Uttar Pradesh. The data was sourced from DACNET and the College of Forestry, SHUATS Prayagraj. The analysis comprised a calibration period of 26 years (90% of the dataset) and a validation period using the remaining data (10%). In this study, 75.9% of the data were utilized for training the Artificial Neural Network (ANN) model, while the remaining 24.1% were allocated for testing and validation, ensuring comprehensive model assessment. The primary evaluation metric employed for model efficiency was the Normalized Root Mean Squared Error (nRMSE), with a focus on achieving the lowest values. Both a Stepwise Linear Regression technique and a Neural Network were employed for rice yield prediction. Notably, the regression-based model exhibited superior performance compared to the ANN model, as indicated by the nRMSE values. This conclusion was drawn from the observation that the regression-based model yielded the best-fitting results. The study's findings highlight the significance of Bright Sunshine Hours in relation to nRMSE and the coefficient of determination, which were recorded at 0.00025 and 0.94, respectively. This underlines the importance of this meteorological factor in accurately predicting rice crop yield.

Impacts of meteorological variables and machine learning algorithms on rice yield prediction in Korea.

As crop productivity is greatly influenced by weather conditions, many attempts have been made to estimate crop yields using meteorological data and have achieved great progress with the development of machine learning. However, most yield prediction models are developed based on observational data, and the utilization of climate model output in yield prediction has been addressed in very few studies. In this study, we estimate rice yields in South Korea using the meteorological variables provided by ERA5 reanalysis data (ERA-O) and its dynamically downscaled data (ERA-DS). After ERA-O and ERA-DS are validated against observations (OBS), two different machine learning models, Support Vector Machine (SVM) and LongShort-Term Memory (LSTM), are trained with different combinations of eight meteorological variables (mean temperature, maximum temperature, minimum temperature, precipitation, diurnal temperature range, solar irradiance, mean wind speed, and relative humidity) obtained from OBS, ERA-O, and ERA-DS at weekly and monthly timescales from May to September. Regardless of the model type and the source of the input data, training a model with weekly datasets leads to better yield estimates compared to monthly datasets. LSTM generally outperforms SVM, especially when the model is trained with ERA-DS data at a weekly timescale. The best yield estimates are produced by the LSTM model trained with all eight variables at a weekly timescale. Altogether this study shows the significance of high spatial and temporal resolution of input meteorological data in yield prediction, which can also serve to substantiate the added value of dynamical downscaling.

Prediction of Rice Yield Based on Multi-Source Data and Hybrid LSSVM Algorithms in China

Prediction of Rice Yield Based on Multi-Source Data and Hybrid LSSVM Algorithms in China

Rice yield prediction using statistical regression models in the selected districts of Maharashtra

Rice yield prediction using statistical regression models in the selected districts of Maharashtra

Grain yield prediction using multi-temporal UAV-based multispectral vegetation indices and endmember abundance in rice

Timely and accurately predicting grain yield before harvest is of great importance for rice production management and grain trade. Numerous methods based on remote sensing (RS) technology have explored to estimate rice grain yield. However, in most cases, these methods are negatively affected by the mixed pixels of RS images due to the effect of background and panicles. To resolve such issues, the abundance information of leaves, soil and water background and rice panicles were extracted from the multispectral images of unmanned aerial vehicle (UAV) for the multiple endmember spectral mixture analysis (MESMA) method. Based on the analysis of contribution of vegetation index (VI) and abundance (ABD) to rice grain yield, a rice grain yield prediction model was developed by combining multi-stage time-series, ABD and VI. Results showed that MESMA can mitigate endmember variability in the estimation of rice yield, and achieve higher accuracy than conventional spectral mixture analysis (SMA). The NDRE and the sum of leaf and panicle abundance (ABDL+P) can be well correlated to grain yield before heading stage (R2 = 0.75) and at heading stage (R2 = 0.72), respectively. In comparison to them, the multi-stage time-series model consisted of VI and ABD [∑(VI&ABD)] produces the highest correlation (R2 = 0.80). It could also resolve the issues about the underestimation of yield using different datasets from various multi-spectral camera in comparison to the single-parameter models ∑VI and ∑ABD, and produce the good validation accuracy with R2 = 0.73, RRMSE= 0.22 and R2 = 0.75, RRMSE= 0.15, respectively. This study suggests that the combination of UAV multi-temporal VI and ABD data can achieve accurate prediction of rice grain yield. This method effectively utilizes the optical information of leaves and rice panicles and reduces background effects, which provides a new idea for accurate rice yield prediction.

Rice crop growth monitoring with sentinel 1 SAR data using machine learning models in google earth engine cloud

The rainfed rice crop monitoring and yield prediction have been Herculean task with optical remote sensing systems operation under cloud cover. The free-of-cost sentinel 1 based SAR data along with machine learning models in GEE cloud was used for rainfed rice crop monitoring for 214 farm plots on a micro-scale in Hooghly, West Bengal, India. The individual plot rice parcel showed a low median backscattering signature for the SAR data during the land preparation/crop transplanting stages with VH and VV at −17.63 dB and −9.63 dB, respectively; whereas, higher median backscattering was experienced at the peak vegetation stage of VH and VV Polarization with −15.20 dB and −6.34 dB, respectively. The random forest model was found best suited with R2 of 0.87 for total crop biomass estimation. The backscatter values have a sound correlation with Heading NDVI, which validated the suitability of SAR images for crop monitoring under rainfed conditions. Further, crop yield prediction using SAR data and total biomass data through machine learning models showed positive correlation for Random forest, Extreme gradient boosting, and Decision tree models with an Area under receiver operating characteristics curve (AUROC) test accuracy of 0.99. These low-cost, high temporal SAR data based models can be used for near real-time crop monitoring even under overcast conditions in near future.

Rice yield prediction using Bayesian analysis on rainfed lands in the Sumbing-Sindoro Toposequence, Indonesia

Since rainfed rice fields typically lack nutrients, frequently experience drought, and require more fund to support farming operations, the production results become erratic and unpredictable. This research aims to construct location-specific rice yield predictions in the rainfed rice fields among the Sumbing-Sindoro Toposequence, Central Java, using a Bayesian method. This study is a survey with an exploratory descriptive methodology based on data from both field and laboratory research. Prediction model analysis using the Bayesian Neural Network (BNN) method on 12geographical units, sampling spots were selected with intention. The following variables were measured: soil (pH level, Organic-C, Total-N, Available-P, Available-K, soil types, elevation, slope) and climate (rainfall, evapotranspiration). According to the statistical analysis used, the BNN model’s performance has the highest accuracy, with an RMSE value of 0.448 t/ha, which compares to the MLR and SR models, indicating the lowest error deviation. To obtain the ideal parameter sampling design, parameter distribution is directly and simultaneously optimised using an optimisation technique based on Pareto optimality. The top 7 data sets (slope, available-P, evapotranspiration, soil type, rainfall, organic-C, and pH) yielded the highest accuracy based on the test results for the three-parameter groups. The coefficient of determination has the highest value, 0.855, while the RMSE test for the model using the top 7 data set has the lowest error value at 0.354 t/ha and 18.71%, respectively. By developing location-specific rice yield predictions using a Bayesian method, farmers and agricultural practitioners can benefit from more accurate and reliable estimates of crop productivity

Development of Rice Yield Forecasting Model Using Linear Regression for Imphal West District, Manipur, India

Rice is a staple food crop and India’s principal food grain. It is generally grown under completely flooded conditions and any changes in weather parameters might affect the rice productivity thereby impacting the food security of the ever-increasing population Prevailing weather conditions during the crop growth period determine the yield of Rice. Hence, the crop yield forecasting models based on weather parameters will be an appropriate option for policymakers and researchers to develop sustainable cropping strategies. The present study examines the application of stepwise multiple linear regression for rice yield prediction using long-term weather data. Analysis was carried out by fixing data from 1998-99 to 2016-17 for calibration and the remaining 2017-18 to 2020-21 data for validation. The accuracy of these models was estimated by R2 (coefficient of determination) and the performance by Mean Square error (MSE), Root mean square error (RMSE), and Normalised root mean square error (NRMSE). The R2 of the developed models ranged from 0.27 – 0.95. The best-performing model was the 5th model with R2 (0.95) with MSE (0.03%), RMSE (0.17%), and NRMSE (0.05%) during validation.

Weather based crop yield prediction using artificial neural networks: A comparative study with other approaches

This paper attempts to compare the weather indices based regression approach and Multilayer Perceptron (MLP) Artificial Neural Network (ANN) approach for rice yield prediction at district level of West Bengal. The weather indices for weather variables, viz., minimum temperature, maximum temperature, rainfall, and relative humidity are used as input variables along with time variable t and yield of rice as output variable. The study reveals that the ANN approach works better than the standard regression approach in crop yield prediction. The prediction error percentages are found to be consistently less than 5% in MLP ANN approach except for one district.