Rapid and accurate prediction of crop nitrogen content is of great significance for guiding precise fertilization. In this study, an unmanned aerial vehicle (UAV) digital camera was used to collect cotton canopy RGB images at 20 m height, and two cotton varieties and six nitrogen gradients were used to predict nitrogen content in the cotton canopy. After image-preprocessing, 46 hand features were extracted, and deep features were extracted by convolutional neural network (CNN). Partial least squares and Pearson were used for feature dimensionality reduction, respectively. Linear regression, support vector machine, and one-dimensional CNN regression models were constructed with manual features as input, and the deep features were used as inputs to construct a two-dimensional CNN regression model to achieve accurate prediction of cotton canopy nitrogen. It was verified that the manual feature and deep feature models constructed from UAV RGB images had good prediction effects. R² = 0.80 and RMSE = 1.67 g kg⁻¹ of the Xinluzao 45 optimal model, and R² = 0.42 and RMSE = 3.13 g kg⁻¹ of the Xinluzao 53 optimal model. The results show that the UAV RGB image and machine learning technology can be used to predict the nitrogen content of large-scale cotton, but due to insufficient data samples, the accuracy and stability of the prediction model still need to be improved.

To realize the classification of sea buckthorn fruits with different water content ranges, a convolution neural network (CNN) detection model of sea buckthorn fruit water content ranges was constructed. In total, 900 images of seabuckthorn fruits with different water contents were collected from 720 seabuckthorn fruits. Eight classic network models based on deep learning were used as feature extraction for transfer learning. A total of 180 images were randomly selected from the images of various water content ranges for testing. Finally, the identification accuracy of the network model for the water content range of seabuckthorn fruit was 98.69%, and the accuracy on the test set was 99.4%. The program in this study can quickly identify the moisture content range of seabuckthorn fruit by collecting images of the appearance and morphology changes during the drying process of seabuckthorn fruit. The model has a good detection effect for seabuckthorn fruits with different moisture content ranges with slight changes in characteristics. The migration deep learning can also be used to detect the moisture content range of other agricultural products, providing technical support for the rapid nondestructive testing of moisture contents of agricultural products.

Soil temperature wireless sensor networks provide essential data for continuous soil temperature monitoring in cotton fields and support agricultural environmental regulation and crop growth management. However, conventional sensor node deployment methods often lead to coverage blind spots, redundant coverage, and insufficient utilization of sensing resources, which restrict network monitoring performance. To address these issues, this study proposes an Adaptive Chaotic Lévy Flight Prairie Dog Optimization algorithm, named ACLFPDO, for optimizing node deployment in soil temperature wireless sensor networks. The incremental novelty of ACLFPDO does not lie in the individual use of chaotic initialization, adaptive parameter adjustment, or Lévy-flight perturbation, which have been widely used in metaheuristic optimization, but in coupling these strategies with a stage-based Prairie Dog Optimization (PDO) position-updating framework and a coverage-oriented fitness design tailored to the STWSN area-coverage problem. An idealized two-dimensional simulation model was established, in which the cotton-field monitoring region was simplified as a regular square area and each sensor node was modeled using a fixed-radius circular binary sensing model. Coverage rate and node utilization efficiency were used as the main evaluation metrics. Comparative simulations were conducted against Prairie Dog Optimization, Snake Optimization (SO), and Whale Optimization algorithms (WO). Under a monitoring area side length of 50, sensing radius of 5, and 42 sensor nodes, ACLFPDO achieved a coverage rate of 98.49% and a node utilization efficiency of 74.65%. Compared with PDO, SO, and WO, the coverage rate increased by 16.88, 9.52, and 10.86 percentage points, respectively. The results indicate that ACLFPDO can improve coverage performance and sensing resource utilization under idealized simulation conditions. However, practical cotton-field deployment still requires further consideration of irregular boundaries, ridges, furrows, obstacles, burial-depth differences, communication connectivity, energy consumption, and soil spatial heterogeneity.