Abstract
BackgroundEstimation of nitrate nitrogen (NO3−–N) content in petioles is one of the key approaches for monitoring nitrogen (N) nutrition in crops. Rapid, non-destructive, and accurate evaluation of NO3−–N contents in cotton petioles under drip irrigation is of great significance.MethodsIn this study, we discussed the use of hyperspectral data to estimate NO3−–N contents in cotton petioles under drip irrigation at different N treatments and growth stages. The correlations among trilateral parameters and six vegetation indices and petiole NO3−–N contents were first investigated, after which a traditional regression model for petioles NO3−–N content was established. A wavelet neural network (WNN) model for estimating petiole NO3−–N content was also established. In addition, the performance of WNN was compared to those of random forest (RF), radial basis function neural network (RBF) and back propagation neural network (BP).ResultsBetween the blue edge amplitude (Db) and blue edge area (SDb) of the blue edge parameters was the optimal index for the estimation model of petiole NO3−–N content. We found that the prediction results of the blue edge parameters and WNN were 7.3% higher than the coefficient of determination (R2) of the first derivative vegetation index and WNN. Root mean square error (RMSE) and mean absolute error (MAE) were 25.2% and 30.9% lower than first derivative vegetation, respectively, and the performance was better than that of RF, RBF and BP.ConclusionsAn inexpensive approach consisting of the WNN algorithm and blue edge parameters can be used to enhance the accuracy of NO3−–N content estimation in cotton petioles under drip irrigation.
Highlights
Estimation of nitrate nitrogen (NO3−–N) content in petioles is one of the key approaches for monitoring nitrogen (N) nutrition in crops
The relationship between NO3−–N contents in petioles and trilateral parameters Table 4 shows that the correlation between N O3−–N content in petioles and blue edge parameters was stronger than that of red edge and yellow edge parameters
The R2 value of the validation model based on blue edge parameters was increased by 7.3%, whereas Root mean square error (RMSE) and mean absolute error (MAE) values were reduced by 25.2% and 30.9%, respectively, when compared to the estimation model based on first derivative vegetation indices
Summary
Estimation of nitrate nitrogen (NO3−–N) content in petioles is one of the key approaches for monitoring nitrogen (N) nutrition in crops. The traditional methods for evaluating cotton N nutrition include soil mineral N determination, laboratory analysis of the plant and determination of petiole NO3−–N levels among others [12, 13]. These methods are associated with certain limitations such as cumbersome procedures, that are time consuming, poor timing of analyses results, and they involve destructive sampling of many plants [14, 15]. Gautam et al [23] used two neural network architectures (Back Propagation and Radial Basis Function) were used to develop twenty different models to predict corn crop N O3−–N content They found that radial basis function model based on green vegetation index textural features provided the best performance with an average accuracy of 92.1%
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have