Farm-Scale Crop Yield Prediction from Multi-Temporal Data Using Deep Hybrid Neural Networks

Farm-scale crop yield prediction is a natural development of sustainable agriculture, producing a rich amount of food without depleting and polluting environmental resources. Recent studies on crop yield production are limited to regional-scale predictions. The regional-scale crop yield predictions usually face challenges in capturing local yield variations based on farm management decisions and the condition of the field. For this research, we identified the need to create a large and reusable farm-scale crop yield production dataset, which could provide precise farm-scale ground-truth prediction targets. Therefore, we utilise multi-temporal data, such as Sentinel-2 satellite images, weather data, farm data, grain delivery data, and cadastre-specific data. We introduce a deep hybrid neural network model to train this multi-temporal data. This model combines the features of convolutional layers and recurrent neural networks to predict farm-scale crop yield production across Norway. The proposed model could efficiently make the target predictions with the mean absolute error of 76 kg per 1000 m2. In conclusion, the reusable farm-scale multi-temporal crop yield dataset and the proposed novel model could meet the actual requirements for the prediction targets in this paper, providing further valuable insights for the research community.

Download Full-text

Crop yield prediction from multi-spectral, multi-temporal remotely sensed imagery using recurrent 3D convolutional neural networks

International Journal of Applied Earth Observation and Geoinformation ◽

10.1016/j.jag.2021.102436 ◽

2021 ◽

Vol 102 ◽

pp. 102436

Author(s):

Mengjia Qiao ◽

Xiaohui He ◽

Xijie Cheng ◽

Panle Li ◽

Haotian Luo ◽

...

Keyword(s):

Neural Networks ◽

Convolutional Neural Networks ◽

Crop Yield ◽

Remotely Sensed ◽

Yield Prediction ◽

Remotely Sensed Imagery ◽

Multi Temporal

Download Full-text

The Implication of Different Sets of Climate Variables on Regional Maize Yield Simulations

Atmosphere ◽

10.3390/atmos11020180 ◽

2020 ◽

Vol 11 (2) ◽

pp. 180

Author(s):

Amit Kumar Srivastava ◽

Andrej Ceglar ◽

Wenzhi Zeng ◽

Thomas Gaiser ◽

Cho Miltin Mboh ◽

...

Keyword(s):

Crop Yield ◽

Large Scale ◽

Regional Scale ◽

Maize Yield ◽

Weather Data ◽

Climate Data ◽

Agricultural Planning ◽

Input Variables ◽

Different Sources ◽

Grid Based

High-resolution and consistent grid-based climate data are important for model-based agricultural planning and farm risk assessment. However, the application of models at the regional scale is constrained by the lack of required high-quality weather data, which may be retrieved from different sources. This can potentially introduce large uncertainties into the crop simulation results. Therefore, in this study, we examined the impacts of grid-based time series of weather variables assembled from the same data source (Approach 1, consistent dataset) and from different sources (Approach 2, combined dataset) on regional scale crop yield simulations in Ghana, Ethiopia and Nigeria. There was less variability in the simulated yield under Approach 1, ranging to 58.2%, 45.6% and 8.2% in Ethiopia, Nigeria and Ghana, respectively, compared to those simulated using datasets retrieved under Approach 2. The two sources of climate data evaluated here were capable of producing both good and poor estimates of average maize yields ranging from lowest RMSE = 0.31 Mg/ha in Nigeria to highest RMSE = 0.78 Mg/ha under Approach 1 in Ghana, whereas, under Approach 2, the RMSE ranged from the lowest value of 0.51 Mg/ha in Nigeria to the highest of 0.72 Mg/ha in Ethiopia under Approach 2. The obtained results suggest that Approach 1 introduces less uncertainty to the yield estimates in large-scale regional simulations, and physical consistency between meteorological input variables is a relevant factor to consider for crop yield simulations under rain-fed conditions.

Download Full-text

Mapping Rice Paddy Based on Machine Learning with Sentinel-2 Multi-Temporal Data: Model Comparison and Transferability

Remote Sensing ◽

10.3390/rs12101620 ◽

2020 ◽

Vol 12 (10) ◽

pp. 1620 ◽

Cited By ~ 1

Author(s):

Weichun Zhang ◽

Hongbin Liu ◽

Wei Wu ◽

Linqing Zhan ◽

Jing Wei

Keyword(s):

Machine Learning ◽

Remote Sensing ◽

Neural Networks ◽

Convolutional Neural Networks ◽

Rice Paddy ◽

Temporal Data ◽

Model Transferability ◽

Local Classification ◽

Multi Temporal ◽

Sentinel 2

Rice is an important agricultural crop in the Southwest Hilly Area, China, but there has been a lack of efficient and accurate monitoring methods in the region. Recently, convolutional neural networks (CNNs) have obtained considerable achievements in the remote sensing community. However, it has not been widely used in mapping a rice paddy, and most studies lack the comparison of classification effectiveness and efficiency between CNNs and other classic machine learning models and their transferability. This study aims to develop various machine learning classification models with remote sensing data for comparing the local accuracy of classifiers and evaluating the transferability of pretrained classifiers. Therefore, two types of experiments were designed: local classification experiments and model transferability experiments. These experiments were conducted using cloud-free Sentinel-2 multi-temporal data in Banan District and Zhongxian County, typical hilly areas of Southwestern China. A pure pixel extraction algorithm was designed based on land-use vector data and a Google Earth Online image. Four convolutional neural network (CNN) algorithms (one-dimensional (Conv-1D), two-dimensional (Conv-2D) and three-dimensional (Conv-3D_1 and Conv-3D_2) convolutional neural networks) were developed and compared with four widely used classifiers (random forest (RF), extreme gradient boosting (XGBoost), support vector machine (SVM) and multilayer perceptron (MLP)). Recall, precision, overall accuracy (OA) and F1 score were applied to evaluate classification accuracy. The results showed that Conv-2D performed best in local classification experiments with OA of 93.14% and F1 score of 0.8552 in Banan District, OA of 92.53% and F1 score of 0.8399 in Zhongxian County. CNN-based models except Conv-1D provided more desirable performance than non-CNN classifiers. Besides, among the non-CNN classifiers, XGBoost received the best result with OA of 89.73% and F1 score of 0.7742 in Banan District, SVM received the best result with OA of 88.57% and F1 score of 0.7538 in Zhongxian County. In model transferability experiments, almost all CNN classifiers had low transferability. RF and XGBoost models have achieved acceptable F1 scores for transfer (RF = 0.6673 and 0.6469, XGBoost = 0.7171 and 0.6709, respectively).

Download Full-text

Satellite-based soybean yield forecast: Integrating machine learning and weather data for improving crop yield prediction in southern Brazil

Agricultural and Forest Meteorology ◽

10.1016/j.agrformet.2019.107886 ◽

2020 ◽

Vol 284 ◽

pp. 107886 ◽

Cited By ~ 12

Author(s):

Raí A. Schwalbert ◽

Telmo Amado ◽

Geomar Corassa ◽

Luan Pierre Pott ◽

P.V.Vara Prasad ◽

...

Keyword(s):

Machine Learning ◽

Crop Yield ◽

Weather Data ◽

Yield Prediction ◽

Southern Brazil ◽

Soybean Yield ◽

Yield Forecast

Download Full-text

Winter Wheat Yield Prediction Using Convolutional Neural Networks from Environmental and Phenological Data

10.21203/rs.3.rs-789462/v1 ◽

2021 ◽

Author(s):

Amit Kumar Srivast ◽

Nima Safaei ◽

Saeed Khaki ◽

Gina Lopez ◽

Wenzhi Zeng ◽

...

Keyword(s):

Machine Learning ◽

Neural Networks ◽

Deep Learning ◽

Winter Wheat ◽

Crop Yield ◽

Wheat Yield ◽

Yield Prediction ◽

Learning Models ◽

Winter Wheat Yield ◽

Machine Learning Models

Abstract Crop yield forecasting depends on many interactive factors including crop genotype, weather, soil, and management practices. This study analyzes the performance of machine learning and deep learning methods for winter wheat yield prediction using extensive datasets of weather, soil, and crop phenology. We propose a convolutional neural network (CNN) which uses the 1-dimentional convolution operation to capture the time dependencies of environmental variables. The proposed CNN, evaluated along with other machine learning models for winter wheat yield prediction in Germany, outperformed all other models tested. To address the seasonality, weekly features were used that explicitly take soil moisture and meteorological events into account. Our results indicated that nonlinear models such as deep learning models and XGboost are more effective in finding the functional relationship between the crop yield and input data compared to linear models and deep neural networks had a higher prediction accuracy than XGboost. One of the main limitations of machine learning models is their black box property. Therefore, we moved beyond prediction and performed feature selection, as it provides key results towards explaining yield prediction (variable importance by time). As such, our study indicates which variables have the most significant effect on winter wheat yield.

Download Full-text