干旱区生态环境知识资源中心(Arid): A comparison of multiple deep learning methods for predicting soil organic carbon in Southern Xinjiang, China

Arid

DOI	10.1016/j.compag.2023.108067
	A comparison of multiple deep learning methods for predicting soil organic carbon in Southern Xinjiang, China
	Wang, Yu; Chen, Songchao; Hong, Yongsheng; Hu, Bifeng; Peng, Jie; Shi, Zhou
通讯作者	Peng, J
来源期刊	COMPUTERS AND ELECTRONICS IN AGRICULTURE
ISSN	0168-1699
EISSN	1872-7107
出版年	2023
卷号	212
英文摘要	Soil organic carbon (SOC) plays an important role in soil functioning and also global C balance. Visible-near -infrared (Vis-NIR) spectroscopy can be regarded as a cost-effective alternative to monitor the SOC content. Previously, application of Vis-NIR spectroscopy in the quantitative estimation of SOC in arid and semi-arid re-gions has received relatively little attention. Here, three different sample sizes of dataset (i.e., 330, 660, and 990) with SOC contents and Vis-NIR spectroscopy measured in the laboratory were obtained from Southern Xinjiang, China. Eight feature selection methods, including Interval Random Frog (IRF), were used to extract the optimal spectral feature subset. Six deep learning (DL) algorithms (e.g., Long Short-Term Memory Neural Networks, LSTM; Deep Belief Networks, DBN) and one machine learning method (Random Forest, RF) were utilized to relate SOC to spectral predictors. The overall objective of this work was to compare the predicted potentials of seven modeling algorithms combined with eight feature selection methods for spectral prediction of SOC. In addition, this paper also investigated the influence of different calibration sample size on the final modeling accuracy for SOC. Results indicated that the DL algorithms outperformed RF for SOC prediction. Among the six DL approaches, the LSTM model performed the best, while the DBN model performed the worst. The one-dimensional-Convolutional Neural Network (1D-CNN), 2D-CNN, Recurrent Neural Network, and DBN algo-rithms were sensitive to different sample sizes. For the largest dataset (i.e., 990 samples), four of the eight feature selection methods combined with the DL algorithms could improve the prediction for SOC, relative to the cor-responding full-spectrum DL models. Among all models developed for SOC, the IRF-LSTM model achieved the optimal prediction, with the validation R2 of 0.89. Our findings provided both theoretical and technical guidance for the spectral estimation of SOC with the relatively low values in arid and semi-arid area.
英文关键词	Deep learning Vis-NIR spectroscopy Soil Organic Carbon Arid region Feature selection
类型	Article
语种	英语
收录类别	SCI-E
WOS记录号	WOS:001047054400001
WOS关键词	OPTIMIZATION ; ALGORITHM ; SELECTION ; SALINITY
WOS类目	Agriculture, Multidisciplinary ; Computer Science, Interdisciplinary Applications
WOS研究方向	Agriculture ; Computer Science
资源类型	期刊论文
条目标识符	http://119.78.100.177/qdio/handle/2XILL650/395817
推荐引用方式 GB/T 7714	Wang, Yu,Chen, Songchao,Hong, Yongsheng,et al. A comparison of multiple deep learning methods for predicting soil organic carbon in Southern Xinjiang, China[J],2023,212.
APA	Wang, Yu,Chen, Songchao,Hong, Yongsheng,Hu, Bifeng,Peng, Jie,&Shi, Zhou.(2023).A comparison of multiple deep learning methods for predicting soil organic carbon in Southern Xinjiang, China.COMPUTERS AND ELECTRONICS IN AGRICULTURE,212.
MLA	Wang, Yu,et al."A comparison of multiple deep learning methods for predicting soil organic carbon in Southern Xinjiang, China".COMPUTERS AND ELECTRONICS IN AGRICULTURE 212(2023).