The nitrogen content of apple leaves and jujube leaves is an important index to judge the growth and development of apple trees and jujube trees to a certain extent. The prediction performance of the two samples was compared between different models for leaf nitrogen content, respectively. The near-infrared absorption spectra of 287 apple leaf samples and 192 jujube leaf samples were collected. After eliminating the outliers by Mahalanobis distance method, the remaining spectral data were processed by six different preprocessing methods. BP neural network (BP), random forest regression (RF), least partial squares (PLS), K-Nearest Neighbor (KNN), and support vector regression (SVR) were compared to establish prediction models of nitrogen content in apple leaves and jujube leaves. The results showed that the determination coefficient (R²), root mean square error (RMSE) and residual prediction deviation (RPD) of the models established by different combined pretreatment methods were compared among the five methods. Compared with the performance of the other four models, the modeling method of SG + SD + CARS + RF was suitable for the prediction of nitrogen content in apple leaves, and its modeling set R² was 0.85408, RMSE was 0.082188, and RPD was 2.5864. The validation set R² is 0.75527, RMSE is 0.099028, RPD is 2.1956. The modeling method of FD + CARS + PLS was suitable for the prediction of nitrogen content in jujube leaves. The modeling set R² was 0.7954, RMSE was 0.14558, and RPD was 2.4264; the validation set R² is 0.81348, RMSE is 0.089217, and RPD is 2.4552.In the prediction modeling of apple leaf nitrogen content in the characteristic band, the model quality of RF was better than the other four prediction models. The model quality of PLS in predictive modeling of nitrogen content of jujube leaves in characteristic bands is superior to the other four predictive models, These results provide a reference for the use of near-infrared spectroscopy to determine whether apple trees and jujube trees are deficient in nutrients.

In order to efficiently detect total nitrogen content in Korla fragrant pear leaves, near-infrared spectroscopy technology was utilized to develop a detection model. The collected spectra underwent various preprocessing techniques including first-order derivative, second-order derivative, Savitzky–Golay + second-order derivative, multivariate scattering correction, multivariate scattering correction + first-order derivative, and standard normal variable transformation + second-order derivative. A competitive adaptive reweighted sampling algorithm was employed to extract characteristic wavelengths, and a prediction model for the total nitrogen content of fragrant pear leaves was established by combining the random forest algorithm, genetic algorithm-based random forest algorithm, radial basis neural network algorithm, and extreme learning machine algorithm. The study found that spectral preprocessing of SNV + SD along with the radial basis neural network algorithm yielded better predictions for total nitrogen content of fragrant pear leaves. The validation set results showed an R² of 0.8547, RMSE of 0.291%, and RPD of 2.699. Therefore, the SNV + SD + CARS + RBF algorithm combination model proved to offer optimal comprehensive performance in predicting the total nitrogen content of fragrant pear leaves.

This study, leveraging near-infrared spectroscopy technology and integrating vegetation index analysis, aims to develop a hyperspectral imaging-based non-destructive inspection technique for swift monitoring of crop chlorophyll content by rapidly predicting leaf SPAD. To this end, a high-precision spectral prediction model was first established under laboratory conditions using ex situ lyophilized Leaf samples. This model provides a core algorithmic foundation for future non-destructive field applications. A systematic study was conducted to develop prediction models for leaf SPAD values of Korla fragrant pear at different growth stages (fruit-setting period, fruit swelling period and Maturity period). This involved comparing various spectral preprocessing algorithms (AirPLS, Savitzky–Golay, Multiplicative Scatter Correction, FD, etc.) and CARS Feature Selection methods for the screening of optimal spectral feature band. Subsequently, models were constructed using BP Neural Network and Support Vector Regression algorithms. The results showed that leaf samples at different growth stages exhibited significant differences in their spectral features within the 5000–7000 cm⁻¹ (effective features for predicting chlorophyll (SPAD)) and 7000–8000 cm⁻¹ (moisture absorption valley) bands. The Savitzky–Golay+FD (Savitzky–Golay smoothing combined with first-order derivative (FD)) preprocessing algorithm performed optimally in feature extraction. Growth period specificity models significantly outperformed whole growth period models, with the optimal models for the fruit-setting period and fruit swelling period being FD-CARS-BP (Coefficient of determination (R²) > 0.86), and the optimal model for the Maturity period being Savitzky–Golay-FD+Savitzky–Golay-CARS-BP (Coefficient_of_determination (R²) = 0.862). Furthermore, joint modeling of characteristic spectra and vegetation indices further improved prediction performance (Coefficient of determination (R²) > 0.85, Root Mean Square Error (RMSE) 2.5). This study presents a reliable method for non-destructive monitoring of chlorophyll content in Korla fragrant pears, offering significant value for nutrient management and stress early warning in precision agriculture.

Leaf total phosphorus content (LTP) is a key indicator for assessing fruit nutrition status. As a rapid non-destructive inspection method, Near-infrared spectroscopy technology is susceptible to the influence of changes in plant growth periods and spectral noise on its prediction accuracy. At present, how to synergistically utilize growth period information and Spectral pre - processing methods to optimize the LTP Prediction model remains to be further studied. The study systematically collected Leaf sample and their near-infrared Spectral data during three key growth periods of Korla fragrant pear (fruit-setting period, fruit swelling period, and Maturity period). In the Spectral pre-processing stage, multiple scattering correction, Savitzky-Golay Smooth, First Derivative (FD), Second Derivative (SD) and their combined algorithms were comprehensively applied. The Competitive Adaptive Reweighted Sampling (CARS) algorithm was used for characteristic wavelength selection, and based on this, Growth period specificity BP neural network model and cross-growth period general prediction models were constructed respectively to evaluate the performance of different Modeling strategies. Results The study showed that LTP content exhibited a significant differential distribution across different growing stage. In the characteristic wavelength bands, after processing with Combined pre-processing method (e.g., MSC+ FD), the correlation coefficient between the spectrum and LTP content significantly increased to approximately 0.90. The predictive performance of the Growth-period-specific model was comprehensively superior to that of the general model, with the Validation set coefficient of determination remaining above 0.83. Compared with the general model, the Coefficient of determination (R²) increased by 0.05-0.16, and the root mean square error decreased by 0.0029-0.0079. This study successfully constructed a technical system of “Growth period-Preprocessing-Model”. The results indicated that the Modeling strategy considering the characteristics of crop growing stage could significantly improve the predictive ability of near-infrared spectroscopy models. This study provides a reliable technical framework for Precision nutrient management in orchard, and the established methodology can also serve as a reference for nutrient Surveillance of other fruit tree plants.

This study investigated the soil–leaf–fruit relationship in Korla fragrant pears (Pyrus sinkiangensis Yu) to establish a scientific cultivation framework by analyzing soil nutrients (alkali-hydrolyzable nitrogen, available phosphorus, available potassium, and pH at 0–60 cm depth) across key phenological stages (fruit setting, expansion, and maturation), combined with leaf and fruit quality indicators. Artificial neural network modeling demonstrated strong predictive capability (R² > 0.85), while sensitivity analysis quantified the relative contributions of different factors, revealing that titratable acidity was optimized when available potassium (30–47 mg/kg) in 40–60 cm soil during fruit setting coincided with pH 7.4–7.8 in 20–40 cm, or when pH 7.3–7.7 in 40–60 cm at fruit setting interacted with alkali-hydrolyzable nitrogen (33.0–53.2 mg/kg) in 40–60 cm during maturation. Fruit shape index improvement required available potassium (40–60 mg/kg) in 40–60 cm at maturation combined with leaf total nitrogen (2.0–6.5 mg/kg) at fruit setting, or specific maturation-stage alkali-hydrolyzable nitrogen levels paired with fruit setting SPAD (Soil and Plant Analysis Development) values (30–41). Furthermore, synergistic effects between expansion stage available phosphorus in 40–60 cm soil and leaf SPAD (Soil and Plant Analysis Development) values simultaneously enhanced the soluble solids content while reducing peel thickness. These findings provide precise nutrient management thresholds for quality optimization, offering practical guidance for orchard management to enhance Korla fragrant pears quality through targeted agricultural practices.

In this study, the effects of natural grass cultivation and clear cultivation on the physiological characteristics of Korla fragrant pear during fruit development and storage were investigated, providing a scientific basis for high-quality fragrant pear cultivation. Sugar components, enzyme activities, and gene expression levels in the pulp and peel were comprehensively analyzed during fruit development and storage. A classification model was constructed using machine learning algorithms (RF, KNN, SVM), and particle swarm optimization (PSO) was employed to identify key factors. The results showed that natural grass cultivation significantly enhanced sugar accumulation in the fruits, particularly at 120 and 150 days after flowering, with fructose content increasing by 9.09 mg·g⁻¹ and 12.59 mg·g⁻¹, respectively, and glucose content also rising significantly. Additionally, natural grass cultivation promoted the relative expression levels of GK, PFK, and FK genes in the pulp. During fruit storage, enzyme activities in the natural grass cultivation group were consistently higher than those in the clear cultivation group across different periods, with PFK activity being 23.73 U/L higher at 150 days of storage. The model identified the activities of glyceraldehyde kinase, phosphofructokinase, and fructokinase as key factors influencing sugar content. A significant negative correlation was observed between peel phosphofructokinase activity and fruit fructose content, though this relationship requires further investigation. This study elucidates the regulatory mechanism by which cultivation methods affect fruit quality through enzyme activity and photosynthetic product distribution. Our findings provide a critical scientific foundation for the high-quality cultivation of Korla fragrant pear and are expected to advance the efficient development of the fragrant pear industry, helping farmers improve both fruit quality and income.