Carrot Yield Mapping: A Precision Agriculture Approach Based on Machine Learning

Carrot yield maps are an essential tool in supporting decision makers in improving their agricultural practices, but they are unconventional and not easy to obtain. The objective was to develop a method to generate a carrot yield map applying a random forest (RF) regression algorithm on a database composed of satellite spectral data and carrot ground-truth yield sampling. Georeferenced carrot yield sampling was carried out and satellite imagery was obtained during crop development. The entire dataset was split into training and test sets. The Gini index was used to find the five most important predictor variables of the model. Statistical parameters used to evaluate model performance were the root mean squared error (RMSE), coefficient of determination (R2) and mean absolute error (MAE). The five most important predictor variables were the near-infrared spectral band at 92 and 79 days after sowing (DAS), green spectral band at 50 DAS and blue spectral band at 92 and 81 DAS. The RF algorithm applied to the entire dataset presented R2, RMSE and MAE values of 0.82, 2.64 Mg ha−1 and 1.74 Mg ha−1, respectively. The method based on RF regression applied to a database composed of spectral bands proved to be accurate and suitable to predict carrot yield.

Download Full-text

A simple approach to estimate coastal soil salinity using digital camera images

Soil Research ◽

10.1071/sr20009 ◽

2020 ◽

Vol 58 (8) ◽

pp. 737

Author(s):

Lu Xu ◽

Raphael A. Viscarra Rossel ◽

Juhwan Lee ◽

Zhichun Wang ◽

Hongyuan Ma

Keyword(s):

Soil Salinity ◽

Precision Agriculture ◽

Mean Squared Error ◽

Rapid Development ◽

Salt Content ◽

Digital Camera ◽

Soil Surface ◽

Coefficient Of Determination ◽

Least Squares Regression ◽

High Concentration

Soil salinisation is a global problem that hinders the sustainable development of ecosystems and agricultural production. Remote and proximal sensing technologies have been used to effectively evaluate soil salinity over large scales, but research on digital camera images is still lacking. In this study, we propose to relate the pixel brightness of soil surface digital images to the soil salinity information. We photographed the surface of 93 soils in the field at different times and weather conditions, and sampled the corresponding soils for laboratory analyses of soil salinity information. Results showed that the pixel digital numbers were related to soil salinity, especially at the intermediate and higher brightness levels. Based on this relationship, we employed random forest (RF) and partial least-squares regression (PLSR) to model soil salt content and ion concentrations, and applied root mean squared error, coefficient of determination and Lin’s concordance correlation coefficient to evaluate the accuracy of models. We found that ions with high concentration were estimated more accurately than ions with low concentrations, and RF models performed overall better than PLSR models. However, the method is only suitable for bare land of coastal soil, and verification is needed for other conditions. In conclusion, a new approach of using digital camera images has good potential to predict and manage soil salinity in the context of precision agriculture with the rapid development of unmanned aerial vehicles.

Download Full-text

Investigation and Optimization of the C-ANN Structure in Predicting the Compressive Strength of Foamed Concrete

Materials ◽

10.3390/ma13051072 ◽

2020 ◽

Vol 13 (5) ◽

pp. 1072 ◽

Cited By ~ 19

Author(s):

Dong Van Dao ◽

Hai-Bang Ly ◽

Huong-Lan Thi Vu ◽

Tien-Thinh Le ◽

Binh Thai Pham

Keyword(s):

Compressive Strength ◽

Civil Engineering ◽

Mean Squared Error ◽

Model Performance ◽

Absolute Error ◽

Machine Learning Algorithms ◽

Coefficient Of Determination ◽

Dry Density ◽

Statistical Measures ◽

Foamed Concrete

Development of Foamed Concrete (FC) and incessant increases in fabrication technology have paved the way for many promising civil engineering applications. Nevertheless, the design of FC requires a large number of experiments to determine the appropriate Compressive Strength (CS). Employment of machine learning algorithms to take advantage of the existing experiments database has been attempted, but model performance can still be improved. In this study, the performance of an Artificial Neural Network (ANN) was fully analyzed to predict the 28 days CS of FC. Monte Carlo simulations (MCS) were used to statistically analyze the convergence of the modeled results under the effect of random sampling strategies and the network structures selected. Various statistical measures such as Coefficient of Determination (R2), Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE) were used for validation of model performance. The results show that ANN is a highly efficient predictor of the CS of FC, achieving a maximum R2 value of 0.976 on the training part and an R2 of 0.972 on the testing part, using the optimized C-ANN-[3–4–5–1] structure, which compares with previous published studies. In addition, a sensitivity analysis using Partial Dependence Plots (PDP) over 1000 MCS was also performed to interpret the relationship between the input parameters and 28 days CS of FC. Dry density was found as the variable with the highest impact to predict the CS of FC. The results presented could facilitate and enhance the use of C-ANN in other civil engineering-related problems.

Download Full-text

Assessment of internal quality attributes of mandarin fruit. 2. NIR calibration model robustness

Australian Journal of Agricultural Research ◽

10.1071/ar04299 ◽

2005 ◽

Vol 56 (4) ◽

pp. 417 ◽

Cited By ~ 36

Author(s):

J. A. Guthrie ◽

D. J. Reid ◽

K. B. Walsh

Keyword(s):

Near Infrared ◽

Model Updating ◽

Multivariate Calibration ◽

Mean Squared Error ◽

Sample Selection ◽

Model Performance ◽

Soluble Solids ◽

Internal Quality ◽

Calibration Model ◽

Validation Population

The robustness of multivariate calibration models, based on near infrared spectroscopy, for the assessment of total soluble solids (TSS) and dry matter (DM) of intact mandarin fruit (Citrus reticulata cv. Imperial) was assessed. TSS calibration model performance was validated in terms of prediction of populations of fruit not in the original population (different harvest days from a single tree, different harvest localities, different harvest seasons). Of these, calibration performance was most affected by validation across seasons (signal to noise statistic on root mean squared error of prediction of 3.8, compared with 20 and 13 for locality and harvest day, respectively). Procedures for sample selection from the validation population for addition to the calibration population (‘model updating’) were considered for both TSS and DM models. Random selection from the validation group worked as well as more sophisticated selection procedures, with approximately 20 samples required. Models that were developed using samples at a range of temperatures were robust in validation for TSS and DM.

Download Full-text

Effects of Subsetting by Carbon Content, Soil Order, and Spectral Classification on Prediction of Soil Total Carbon with Diffuse Reflectance Spectroscopy

Applied and Environmental Soil Science ◽

10.1155/2012/294121 ◽

2012 ◽

Vol 2012 ◽

pp. 1-14 ◽

Cited By ~ 20

Author(s):

Meryl L. McDowell ◽

Gregory L. Bruland ◽

Jonathan L. Deenik ◽

Sabine Grunwald

Keyword(s):

Diffuse Reflectance ◽

Near Infrared ◽

Prediction Models ◽

Mean Squared Error ◽

Diffuse Reflectance Spectroscopy ◽

Reflectance Spectroscopy ◽

Total Carbon ◽

Coefficient Of Determination ◽

Spectral Classification ◽

Sample Set

Subsetting of samples is a promising avenue of research for the continued improvement of prediction models for soil properties with diffuse reflectance spectroscopy. This study examined the effects of subsetting by soil total carbon (Ct) content, soil order, and spectral classification withk-means cluster analysis on visible/near-infrared and mid-infrared partial least squares models forCtprediction. Our sample set was composed of various Hawaiian soils from primarily agricultural lands withCtcontents from <1% to 56%. Slight improvements in the coefficient of determination (R2) and other standard model quality parameters were observed in the models for the subset of the high activity clay soil orders compared to the models of the full sample set. The other subset models explored did not exhibit improvement across all parameters. Models created from subsets consisting of only lowCtsamples (e.g.,Ct< 10%) showed improvement in the root mean squared error (RMSE) and percent error of prediction for lowCtsoil samples. These results provide a basis for future study of practical subsetting strategies for soilCtprediction.

Download Full-text

Modeling Wood Density of Larch by Near-Infrared Spectrometry and Support Vector Machine

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.55-57.433 ◽

2011 ◽

Vol 55-57 ◽

pp. 433-438

Author(s):

Ya Zhao Zhang ◽

Yao Xiang Li ◽

Hong Fu Zhang ◽

Hui Juan Zhang ◽

Pai Li

Keyword(s):

Support Vector Machine ◽

Infrared Spectroscopy ◽

Wood Density ◽

Near Infrared Spectroscopy ◽

Near Infrared ◽

Model Building ◽

Model Performance ◽

Coefficient Of Determination ◽

Infrared Spectrometry ◽

Support Vector

Model for predicting wood density of Larch was established using near-infrared spectroscopy (NIR) combined with support vector machine (SVM). A hundred and seventeen Larch samples were used in the study. Wood density of samples was measured according to standard test methods for physical and mechanical properties of wood. Support vector machines for regression (SVR) was used for model building. Radial basis function (RBF) was used as kernel function to establish a model for predicting wood density. For the train set, the coefficient of determination (R2) and the mean square error (MSE) were 0.8504 and 0.6460×10-3, while the R2 and MSE was 0.8520 and 0.4451×10-3, respectively, for the test set. Results showed that using SVM in near-infrared spectroscopy calibration could significantly improve the model performance in order to rapidly and accurately predict wood density.

Download Full-text

Canopy Temperature as a Key Physiological Trait to Improve Yield Prediction under Water Restrictions in Potato

Agronomy ◽

10.3390/agronomy11071436 ◽

2021 ◽

Vol 11 (7) ◽

pp. 1436

Author(s):

Johan Ninanya ◽

David A. Ramírez ◽

Javier Rinza ◽

Cecilia Silva-Díaz ◽

Marcelo Cervantes ◽

...

Keyword(s):

Decision Support ◽

Stomatal Conductance ◽

Mean Squared Error ◽

Satellite Image ◽

Model Performance ◽

Canopy Temperature ◽

Coefficient Of Determination ◽

Yield Prediction ◽

Crop Models ◽

Water Treatments

Canopy temperature (CT) as a surrogate of stomatal conductance has been highlighted as an essential physiological indicator for optimizing irrigation timing in potatoes. However, assessing how this trait could help improve yield prediction will help develop future decision support tools. In this study, the incorporation of CT minus air temperature (dT) in a simple ecophysiological model was analyzed in three trials between 2017 and 2018, testing three water treatments under drip (DI) and furrow (FI) irrigations. Water treatments consisted of control (irrigated until field capacity) and two-timing irrigation based on physiological thresholds (CT and stomatal conductance). Two model perspectives were implemented based on soil water balance (P1) and using dT as the penalizing factor (P2), affecting the biomass dynamics and radiation use efficiency parameters. One of the trials was used for model calibration and the other two for validation. Statistical indicators of the model performance determined a better yield prediction at harvest for P2, especially under maximum stress conditions. The P1 and P2 perspectives showed their highest coefficient of determination (R2) and lowest root-mean-squared error (RMSE) under DI and FI, respectively. In the future, the incorporation of CT combining low-cost infrared devices/sensors with spatial crop models, satellite image information, and telemetry technologies, an adequate decision support system could be implemented for water requirement determination and yield prediction in potatoes.

Download Full-text

Modelling Reservoir Turbidity Using Landsat 8 Satellite Imagery by Gene Expression Programming

Water ◽

10.3390/w11071479 ◽

2019 ◽

Vol 11 (7) ◽

pp. 1479 ◽

Cited By ~ 7

Author(s):

Liu ◽

Wang

Keyword(s):

Gene Expression ◽

Performance Evaluation ◽

Satellite Imagery ◽

Mean Squared Error ◽

Accuracy Assessment ◽

Gene Expression Programming ◽

Model Performance ◽

Coefficient Of Determination ◽

Landsat 8 ◽

Model Performance Evaluation

This study aimed to develop a reliable turbidity model to assess reservoir turbidity based on Landsat-8 satellite imagery. Models were established by multiple linear regression (MLR) and gene-expression programming (GEP) algorithms. Totally 55 and 18 measured turbidity data from Tseng-Wen and Nan-Hwa reservoir paired and screened with satellite imagery. Finally, MLR and GEP were applied to simulated 13 turbid water data for critical turbidity assessment. The coefficient of determination (R2), root mean squared error (RMSE), and relative RMSE (R-RMSE) calculated for model performance evaluation. The result show that, in model development, MLR and GEP shows a similar consequent. However, in model testing, the R2, RMSE, and R-RMSE of MLR and GEP are 0.7277 and 0.8278, 0.7248 NTU and 0.5815 NTU, 22.26% and 17.86%, respectively. Accuracy assessment result shows that GEP is more reasonable than MLR, even in critical turbidity situation, GEP is more convincible. In the model performance evaluation, MLR and GEP are normal and good level, in critical turbidity condition, GEP even belongs to outstanding level. These results exhibit GEP denotes rationality and with relatively good applicability for turbidity simulation. From this study, one can conclude that GEP is suitable for turbidity modeling and is accurate enough for reservoir turbidity estimation.

Download Full-text

Calibration and validation of the AquaCrop model for the soybean crop grown under different levels of irrigation in the Motopiba region, Brazil

Ciência Rural ◽

10.1590/0103-8478cr20161118 ◽

2017 ◽

Vol 48 (1) ◽

Cited By ~ 2

Author(s):

Vicente de Paulo Rodrigues da Silva ◽

Roberta Araújo e Silva ◽

Girlene Figueiredo Maciel ◽

Célia Campos Braga ◽

José Luiz Cabral da Silva Júnior ◽

...

Keyword(s):

Model Performance ◽

Absolute Error ◽

Efficiency Index ◽

Water Levels ◽

Coefficient Of Determination ◽

Yield Response ◽

Statistical Parameters ◽

Yield Data ◽

Climate Conditions ◽

Aquacrop Model

ABSTRACT: The water-driven AquaCrop model to simulate yield response has been calibrated and validated for soybean cultivated under different water levels irrigation in Matopiba region, Brazil. The crop was submitted to seven irrigation treatments during the dry season and a dry treatment in the rainy season. The model was parameterized and calibrated by using soybean yield data collected at field level. Model performance was evaluated by using the following statistical parameters: prediction error (Pe), Nash-Sutcliffe efficiency index (E), coefficient of determination (R2), mean absolute error (MAE), root mean square error normalized (RMSEN) and Willmott’s index (d). The statistical analyses of the AquaCrop model calibrated for the Matopiba region disclosed error acceptable for yield prediction of soybean grown under tropical climate conditions. Results also indicated that the C2 soybean cultivar is more resistant to water stress than the C1 soybean grown in the Matopiba region, Brazil. In the treatments when the crop was well supplied with water, at least in one phase, the yield was greater than those with drought stress at last in one phase.

Download Full-text

Alfalfa Yield Prediction Using UAV-Based Hyperspectral Imagery and Ensemble Learning

Remote Sensing ◽

10.3390/rs12122028 ◽

2020 ◽

Vol 12 (12) ◽

pp. 2028 ◽

Cited By ~ 3

Author(s):

Luwei Feng ◽

Zhou Zhang ◽

Yuchi Ma ◽

Qingyun Du ◽

Parker Williams ◽

...

Keyword(s):

Precision Agriculture ◽

Vegetation Indices ◽

Model Performance ◽

Original Data ◽

The United States ◽

Hyperspectral Data ◽

Coefficient Of Determination ◽

Support Vector ◽

Yield Prediction ◽

Ensemble Model

Alfalfa is a valuable and intensively produced forage crop in the United States, and the timely estimation of its yield can inform precision management decisions. However, traditional yield assessment approaches are laborious and time-consuming, and thus hinder the acquisition of timely information at the field scale. Recently, unmanned aerial vehicles (UAVs) have gained significant attention in precision agriculture due to their efficiency in data acquisition. In addition, compared with other imaging modalities, hyperspectral data can offer higher spectral fidelity for constructing narrow-band vegetation indices which are of great importance in yield modeling. In this study, we performed an in-season alfalfa yield prediction using UAV-based hyperspectral images. Specifically, we firstly extracted a large number of hyperspectral indices from the original data and performed a feature selection to reduce the data dimensionality. Then, an ensemble machine learning model was developed by combining three widely used base learners including random forest (RF), support vector regression (SVR) and K-nearest neighbors (KNN). The model performance was evaluated on experimental fields in Wisconsin. Our results showed that the ensemble model outperformed all the base learners and a coefficient of determination (R2) of 0.874 was achieved when using the selected features. In addition, we also evaluated the model adaptability on different machinery compaction treatments, and the results further demonstrate the efficacy of the proposed ensemble model.

Download Full-text

Predicting plant available phosphorus using infrared spectroscopy with consideration for future mobile sensing applications in precision farming

Precision Agriculture ◽

10.1007/s11119-019-09693-3 ◽

2019 ◽

Vol 21 (4) ◽

pp. 737-761 ◽

Cited By ~ 4

Author(s):

Stefan Pätzold ◽

Matthias Leenen ◽

Peter Frizen ◽

Tobias Heggemann ◽

Peter Wagner ◽

...

Keyword(s):

Precision Agriculture ◽

Near Infrared ◽

Model Performance ◽

Precision Farming ◽

Future Application ◽

Available Phosphorus ◽

Agricultural Practice ◽

Mir Spectroscopy ◽

Sensing Applications ◽

Rapid Discrimination

Abstract Phosphorus (P) fertilisation recommendations rely primarily on soil content of plant available P (Pavl) that vary spatially within farm fields. Spatially optimized P fertilisation for precision farming requires reliable, rapid and non-invasive Pavl determination. This laboratory study aimed to test and to compare visible-near infrared (Vis–NIR) and mid-infrared (MIR) spectroscopy for Pavl prediction with emphasis on future application in precision agriculture. After calibration with the conventional calcium acetate lactate (CAL) extraction method, limitations of Vis–NIRS and MIRS to predict Pavl were evaluated in loess topsoil samples from different fields at six localities. Overall calibration with 477 (Vis–NIRS) and 586 (MIRS) samples yielded satisfactory model performance (R2 0.70 and 0.72; RPD 1.8 and 1.9, respectively). Local Vis–NIRS models yielded better results with R2 up to 0.93 and RPD up to 3.8. For MIRS, results were comparable. However, an overall model to predict Pavl on independent test data partly failed. Sampling date, pre-crop harvest residues and fertilising regime affected model transferability. Varying transferability could partly be explained after deriving the cellulose absorption index from the Vis–NIR spectra. In 62 (Vis–NIRS) and 67% (MIRS) of all samples, prediction matched the correct Pavl content class. Rapid discrimination between high, optimal and low P classes could be carried out on many samples from single fields thus marking an improvement over the common practice. However, Pavl determination by means of IR spectroscopy is not yet satisfactory for determination of precision fertilizer dosage. For introduction into agricultural practice, a standardized sampling protocol is recommended to help achieve reliable spectroscopic Pavl prediction.

Download Full-text