褐土全氮含量Vis/NIRS组合预测模型的构建

doi:10.3964/j.issn.1000-0593(2024)08-2310-08

Abstract
References
Related Citation (15)

Download: PDF (5249 KB)
Export: BibTeX | EndNote (RIS)

Abstract Accurately grasping the total nitrogen content of farmland soil is significant for evaluating soil fertility and applying nitrogen fertilizer reasonably. To comprehensively utilize the advantages of each single prediction Model, improve the overall prediction performance, reduce the variance of the model, and improve the robustness, this study takes farmland brown soil as the research object, and based on near-infrared and visible hyperspectral data, puts forward a Combined prediction model based on standard deviation. CPM was used to predict soil total nitrogen content. Savitzky-Golay smoothing and first-order differential transformation are applied to the original hyperspectral data, and a tree model is used for feature band extraction. Using five single prediction models, Decision Tree Regression (DTR) (Model 1), Gaussian Kernel Regression (GKR) (Model 2), Random Forest Regression (RF) (Model 3), LASSO Regression (Model 4), and Multi-Layer Perceptron (MLP) (Model 5), a combination prediction model is established through a linear combination of single prediction models. The results indicate that: (1) The weights of the five single prediction models in the combined prediction model are obtained by generalized reduced gradient optimization algorithm: ω^*₁=0.407，ω^*₂=0.378，ω^*₃=0.215，ω^*₄=0，ω^*₅=0; (2) For all data, the predictive effectiveness of five single prediction models and combined prediction models for predicting soil total nitrogen content is M, respectively M₁=0.855，M₂=0.856，M₃=0.847，M₄=0.785，M₅=0.796，M_CPM=0.880, compared to the maximum predictive validity of a single model, the predictive validity of the combination prediction model has increased by 2.924%; (3) For all data, the prediction accuracy and standard deviation of soil total nitrogen content based on five single prediction models and combined prediction models are E(A₁)=0.924，σ(A₁)=0.075，E(A₂)=0.928，σ(A₂)=0.077，E(A₃)=0.923，σ(A₃)=0.082，E(A₄)=0.882，σ(A₄)=0.109，E(A₅)=0.889，σ(A₅)=0.104，E(A_CPM)=0.937，σ(A_CPM)=0.066, compared to the maximum prediction accuracy of a single model, the combination prediction model has improved prediction accuracy by 0.970% and model stability by 12.000%, making it an optimal combination prediction model. The combined prediction model can effectively estimate the total nitrogen content of farmland brown soil based on visible-near-infrared spectral data and can provide a basis and reference for the rapid monitoring of the total nitrogen content of farmland soil.

Key words：Visible-near infrared; Hyperspectral estimation; Total nitrogen content; Combination prediction model

Received: 2023-04-17 Accepted: 2023-09-21

ZTFLH:

S153.621

Corresponding Authors: LI Zhi-wei E-mail: lizhiweitong@163.com

Cite this article:

ZHANG Xiu-quan,MA Shi-xing,LI Zhi-wei, et al. Prediction of Total Nitrogen Content in Brown Soil Based on Hyperspectral and Combined Prediction Model[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(08): 2310-2317.

URL:

https://www.gpxygpfx.com/EN/10.3964/j.issn.1000-0593(2024)08-2310-08 OR https://www.gpxygpfx.com/EN/Y2024/V44/I08/2310

[1] Liu Jun, Cai Haotian, Chen Shan, et al. Agriculture, 2023, 13（4）：743.
[2] ZHOU Yang, ZHAO Xiao-min, GUO Xi（周洋, 赵小敏, 郭熙）. Acta Pedologica Sinica（土壤学报）, 2022, 59（2）：451.
[3] Bai Zijin, Xie Modong, Hu Bifeng, et al. Sensors, 2022, 22（16）：6124.
[4] ZHANG Yao, CUI Yun-tian, DENG Qiu-zhuo, et al（张瑶, 崔云天, 邓秋卓, 等）. Transactions of the Chinese Society for Agricultural Machinery（农业机械学报），2021，52（S1）：310.
[5] CHEN Wei, XU Zhan-jun, GUO Qi(陈玮，徐占军，郭琦). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2022, 38（8）：9.
[6] Zhang Xianglin, Xue Jie, Xiao Yi, et al. Remote Sensing, 2023, 15（2）：465.
[7] Lin Lixin, Liu Xixi. Computer and Electronics in Agriculture, 2022, 192（0）：106634.
[8] ZHANG Xin, YANG Chao, LIU Hong-bin, et al（张鑫, 杨超, 刘洪斌, 等）. Tobacco Science & Technology（烟草科技）, 2022, 55（8）：20.
[9] CHEN Hao-yu, XIANG Lei, GAO He, et al（陈昊宇, 项磊, 高贺, 等）. Remote Sensing for Natural Resources（自然资源遥感）, 2023, 35（3）：170.
[10] YANG Mei-hua, ZHAO Xiao-min, FANG Qian, et al（杨梅花, 赵晓敏, 方倩, 等）. Scientia Agricultura Sinica（中国农业科学）, 2014, 47（12）：2374.
[11] Shi T, Cui L, Wang J, et al. Plant and Soil, 2013，366：313.
[12] Wang Y, Li M, Ji R, et al. Sensors, 2020, 20（24）：7078.
[13] Ma J J, Cheng J, Wang J H, et al. Information Processing in Agriculture, 2022, 9: 566.
[14] ZHANG Heng, LIANG Tai-bo, FENG Wen-qiang, et al（张恒, 梁太波, 冯文强, 等）. Chinese Tobacco Science（中国烟草科学）, 2023, 44（5）：103.
[15] ZHANG Xiu-quan, LI ZHi-wei, ZHENG De-cong, et al（张秀全, 李志伟, 郑德聪, 等）. Spectroscopy and Spectral Analysis（光谱学与光谱分析）, 2023, 43（3）：903.
[16] ZHOU Zhi-hua（周志华）. Maching Learning（机器学习）. Beijing: Tsinghua University Press（北京：清华大学出版社）, 2016.
[17] DONG Xue-mei, WANG Jie-wei（董雪梅, 王洁微）. Pattern Recognition and Artificial Intelligence（模式识别与人工智能）, 2019, 32（7）：589.
[18] WANG Lu, SUN Ju-bo（王璐, 孙聚波）. Journal of Jilin Engineering Normal University（吉林工程技术师范学院学报）, 2021, 37（12）：109.
[19] WANG Qiang, HU Rong（王强, 胡荣）. Computer Engineering and Design（计算机工程与设计）, 2022, 43（12）：3443.
[20] CHEN Hua-you, HOU Ding-pi（陈华友，侯定丕）. Journal of Systems Engineering（系统工程学报）, 2003, 18（3）：203.
[21] MA Wei-biao, WANG Hong-yan, RUI Qiang(马伟标，王红岩，芮强). Journal of System Simulation(系统仿真学报)，2012，24(4)：774.
[22] Stenberg B, Viscarra Rossel R A, Mouazen A M. Advances in Agronomy, 2010, 107：163.
[23] WANG Wen-cai, ZHAO Liu, LI Shao-wen, et al（王文才，赵刘，李绍稳，等）. Acta Agricultruae Zhejiangensis（浙江农业学报）, 2018, 30（9）：1576.
[24] Wang Y, Liao Z, Stéphanie Mathieu, et al. Journal of Hazardous Materials, 2021, 404(Pt A)：123965.
[25] ZHANG Peng-cheng, LI Wei-juan, WANG Jun（张朋程, 李伟娟, 王军）. China Energy and Environmental Protection（能源与环保）, 2023, 45（6）：198.
[26] Hong Y, Liu Y, Chen Y, et al. Geoderma, 2019, 337：758.