Prediction Method of Available Nitrogen in Red Soil Based on BWO-CNN-LSTM
Source abstract
Abstract
Accurate assessment of forest soil nitrogen from hyperspectral spectra is critical for precision fertilization, yet conventional preprocessing and baseline CNNs constrain predictive accuracy. We introduce streamlined spectral preprocessing and an optimized CNN–LSTM framework and evaluate it on Guangxi forest soils against competitive models using standard validation metrics. Results: The proposed approach outperformed comparative models (CNN, LSTM, and BiLSTM), achieving a validation set R2 of 0.889 and RMSE of 16.5722, representing improvements of 6.79–10.37% in R2 and 18.60–24.44% in RMSE over baseline methods. The method delivers accurate, scalable nitrogen estimation from spectra, supporting timely fertilization decisions and sustainable soil management.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Agriculture, Soil, Plant Science · Remote Sensing and Land Use · Image Processing and 3D Reconstruction
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Computer Science and Engineering/School of Software, Guangxi Normal University, Guilin 541006, China
College of Computer Science and Engineering, Guilin University of Technology, Guilin 541006, China
Guangxi Key Laboratory of Embedded Technology and Intelligent System, Guilin University of Technology, Guilin 541006, China
Funding
Guangxi Key Research and Development Program · National Natural Science Foundation of China · Innovation Project of Guangxi Graduate Education
References (24)
Show source-provided references
1. Prediction of soil total nitrogen content based on CatBoost algorithm and graph feature fusion
2. Hyperspectral estimation modeling of soil As content in the eastern part of Tianfu New District, Chengdu based on SPA-BPNN
3. Indirect estimation of organic matter in cultivated soil based on surface soil spectrum
4. Nitrogen-total
5. Prediction of soil organic carbon for different levels of soil moisture using Vis-NIR spectroscopy10.1016/j.geoderma.2012.07.020 ↗
6. Visible, near-infrared, and mid-infrared spectroscopy applications for soil assessment with emphasis on soil organic matter content and quality: State-of-the-art and key issues10.1366/13-07288 ↗
7. Hyperspectral estimation of organic matter in red soil based on different convolutional neural network models
8. Study on hyperspectral model of organic matter content in black soil
9. Inversion of total nitrogen content in rubber plantation soil in Yunnan based on hyperspectral analysis
10. Comparative study on soil total nitrogen content estimation models based on hyperspectral data
11. Study on hyperspectral model of total nitrogen and total phosphorus content in wetland soil
12. Kim, S., Choi, H.-J., Park, S., and Kim, Y. (2025). Predicting the Freezing Characteristics of Organic Soils Using Laboratory Experiments and Machine Learning Models. Appl. Sci., 15.10.3390/app151910314 ↗
13. Geng, W., Li, T., Zhu, X., Dou, L., Liu, Z., Qian, K., Ye, G., Lin, K., Li, B., and Ma, X. (2025). Predicting the Zinc Content in Rice from Farmland Using Machine Learning Models: Insights from Universal Geochemical Parameters. Appl. Sci., 15.10.3390/app15031273 ↗
14. Alahmad, T., Neményi, M., and Nyéki, A. (2025). Soil Moisture Content Prediction Using Gradient Boosting Regressor (GBR) Model: Soil-Specific Modeling with Five Depths. Appl. Sci., 15.10.3390/app15115889 ↗
15. Ryu, S., Kim, J., Choi, H., Lee, J., and Han, J. (2025). Case Study on Analysis of Soil Compression Index Prediction Performance Using Linear and Regularized Linear Machine Learning Models (In Korea). Appl. Sci., 15.10.3390/app15052757 ↗
16. Quantitative modeling method of corn components based on near-infrared spectral fusion and deep learning
17. Remote sensing fine mapping of agricultural planting structure based on spatiotemporal coordination10.11834/jrs.20241349 ↗
18. Dataset representativeness measurement and block sampling strategy for the pan-Kennard-Stone algorithm
19. Estimation of soil organic matter content in mountainous cultivated land based on hyperspectral analysis
20. Deep convolutional neural networks for hyperspectral image classification10.1155/2015/258619 ↗
21. Ideal spatial adaptation by wavelet shrinkage10.1093/biomet/81.3.425 ↗
22. Beluga whale optimization: A novel nature-inspired metaheuristic algorithm10.1016/j.knosys.2022.109215 ↗
23. He, J., Li, S., Shen, J., Liu, Y., Wang, J., and Jin, P. (December, January 30). Facial expression recognition based on VGGNet convolutional neural network. Proceedings of the 2018 Chinese Automation Congress (CAC), Xi’an, China.
24. Targ, S., and Almeida, D. (2016). Resnet in resnet: Generalizing residual architectures. arXiv.