Abstract
Leaf area index (LAI) is a crucial crop biophysical parameter that has been widely used in a variety of fields. Five state-of-the-art machine learning regression algorithms (MLRAs), namely, artificial neural network (ANN), support vector regression (SVR), Gaussian process regression (GPR), random forest (RF) and gradient boosting regression tree (GBRT), have been used in the retrieval of cotton LAI with Sentinel-2 spectral bands. The performances of the five machine learning models are compared for better applications of MLRAs in remote sensing, since challenging problems remain in the selection of MLRAs for crop LAI retrieval, as well as the decision as to the optimal number for the training sample size and spectral bands to different MLRAs. A comprehensive evaluation was employed with respect to model accuracy, computational efficiency, sensitivity to training sample size and sensitivity to spectral bands. We conducted the comparison of five MLRAs in an agricultural area of Northwest China over three cotton seasons with the corresponding field campaigns for modeling and validation. Results show that the GBRT model outperforms the other models with respect to model accuracy in average ( R 2 ¯ = 0.854, R M S E ¯ = 0.674 and M A E ¯ = 0.456). SVR achieves the best performance in computational efficiency, which means it is fast to train, and to validate that it has great potentials to deliver near-real-time operational products for crop management. As for sensitivity to training sample size, GBRT behaves as the most robust model, and provides the best model accuracy on the average among the variations of training sample size, compared with other models ( R 2 ¯ = 0.884, R M S E ¯ = 0.615 and M A E ¯ = 0.452). Spectral bands sensitivity analysis with dCor (distance correlation), combined with the backward elimination approach, indicates that SVR, GPR and RF provide relatively robust performance to the spectral bands, while ANN outperforms the other models in terms of model accuracy on the average among the reduction of spectral bands ( R 2 ¯ = 0.881, R M S E ¯ = 0.625 and M A E ¯ = 0.480). A comprehensive evaluation indicates that GBRT is an appealing alternative for cotton LAI retrieval, except for its computational efficiency. Despite the different performance of the ML models, all models exhibited considerable potential for cotton LAI retrieval, which could offer accurate crop parameters information timely and accurately for crop fields management and agricultural production decisions.
Highlights
Leaf area index (LAI), which characterizes the structure and functioning of vegetation, is usually defined as half of the total green leaf area per unit horizontal ground surface area [1,2]
To identify the performance of five popular machine learning (ML) algorithms for cotton LAI retrieval, regression models were established based on artificial neural network (ANN), support vector regression (SVR), Gaussian process regression (GPR), random forest (RF) and gradient boosting regression tree (GBRT), at S2-20 m for the whole growth period of cotton
To evaluate the performance of the ML regression models, the root mean square error (RMSE), mean absolute error (MAE) and coefficient of determination (R2 ) between the measured and predicted values were used to assess the performance of the models
Summary
Leaf area index (LAI), which characterizes the structure and functioning of vegetation, is usually defined as half of the total green leaf area per unit horizontal ground surface area [1,2]. Remote sensing techniques provide promising alternatives to obtaining crop biophysical parameters by high temporally and spatially continuous means over large areas. Parametric regression methods usually consist of an explicit relationship between biophysical parameters and vegetation indices, while non-parametric regression methods define regression models learnt from the training dataset [15]. While physically-based methods are applications of physical laws establishing cause-effect relationships, a hybrid method combines elements of non-parametric statistics and physically-based methods [13], whereas these two methods are both sophisticated models that demand a large number of parameters, which are usually difficult to obtain in practice. Nonparametric models can make full use of spectral information, and directly learn the input-output relationships from a given training dataset, which makes these models attractive alternatives for crop LAI retrieval
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
