干旱区生态环境知识资源中心(Arid): 基于竞争适应重加权采样算法耦合机器学习的土壤含水量估算

Arid

	基于竞争适应重加权采样算法耦合机器学习的土壤含水量估算
其他题名	Estimation of Soil Moisture Content Based on Competitive Adaptive Reweighted SamplingAlgorithm Coupled with Machine Learning
	葛翔宇 1; 丁建丽1 ; 王敬哲 1; 王飞 1; 蔡亮红 1; 孙慧兰 2
来源期刊	光学学报
ISSN	0253-2239
出版年	2018
卷号	38 期号:10 页码:1030001-1-1030001-8
中文摘要	土壤含水量是干旱区地表水-热-溶质耦合运移的关键指标;以干旱区典型样点实测土壤含水量及其室内可见光-近红外光谱数据作为数据集,通过蒙特卡罗交叉验证确定77个有效样本;基于竞争适应重加权采样算法筛选出最优光谱变量子集,利用3种机器学习方法BP神经网络、随机森林回归和极限学习机建立土壤含水量预测模型,进而实现土壤含水量估算模型的优选。结果表明:竞争适应重加权采样算法能有效剔除无关变量,从2151个光谱波段中优选出20个特征波段,其中R_(1848)与土壤含水量的最大相关系数为0.531;引入偏最小二乘模型和机器学习方法进行对比,分析发现机器学习方法的预测结果比偏最小二乘模型更高;分析比较BP神经网络、随机森林回归和极限学习机的建模结果可知:极限学习机模型建模在机器学习方法中的效果最佳,决定系数R~2=0.918,均方根误差RMSE=0.015,相对分析误差RPD=3.123,四分位数间隔RPIQ=3.325;机器学习能显著提升光谱建模反演土壤含水量的精度和稳定性,显示出其在非线性问题中具有很强的透析力和较好的模型稳健性,针对干旱区土壤水分的精准预测和定量估算具有可行性,可为干旱区土壤墒情、精准农业等研究提供科学参考。
英文摘要	Soil moisture content is an important indicator that reflects the coupled surface water-heat-solute transport in arid regions.The visible and near-infrared spectroscopy has been widely used for soil moisture content prediction owing to its rapid response.The soil moisture content and corresponding spectral data are obtained in the laboratory;then,the calibration datasets(n=77)are selected using Monte Carlo cross-validation algorithm.The competitive adaptive reweighted sampling algorithm is used to optimize spectral variables.Three machine learning algorithms,namely back propagation neural network,random forest regression,and extreme learning machine are used to construct predicting models.The results reveal that competitive adaptive reweighted sampling algorithm can effectively filter and eliminate massive irrelevant variables.Herein,a total of 20 feature bands are divided from all spectral bands,where the band of R_(1848) is the most prominent(the maximum correlation coefficient is 0.531).The performance of models based on machine learning algorithms is superior to those based on partial least squares regression,with the optimal prediction of the coefficient of determination (R~2),root mean square error of prediction(RMSE),residual predictive deviation(RPD),and ratio of performance to interquartile range(RPIQ).Compared with the predictive effects of all the models,the extreme learning machine-based predicting model is the most effective(R~2=0.918,RMSE=0.015,RPD=3.123,and RPIQ=3.325).Compared with common linear models,the machine learning algorithms can effectively improve the precision and stability of the quantitative estimation of soil moisture content.The results provide scientific guidance and baseline data for the accurate monitoring of soil moisture content and precision agriculture in arid regions.
中文关键词	光谱学 ; 土壤含水量估算 ; 机器学习 ; 竞争适应重加权采样算法 ; 极限学习机 ; 随机森林
英文关键词	spectroscopy estimation of soil moisture content machine learning competitive adaptive reweighted sampling algorithm extreme learning machine random forest
语种	中文
国家	中国
收录类别	CSCD
WOS类目	PHYSICS MULTIDISCIPLINARY ; OPTICS
WOS研究方向	Physics ; Optics
CSCD记录号	CSCD:6354647
来源机构	新疆师范大学 ; 新疆大学
资源类型	期刊论文
条目标识符	http://119.78.100.177/qdio/handle/2XILL650/237798
作者单位	1.新疆大学资源与环境科学学院;;新疆大学;;新疆大学, ;;绿洲生态教育部重点实验室;;智慧城市与环境建模自治区普通高校重点实验室, 乌鲁木齐;;乌鲁木齐;;乌鲁木齐, 新疆;;新疆;;新疆 830046;;830046;;830046, 中国; 2.新疆师范大学地理科学与旅游学院, 乌鲁木齐, 新疆 830054, 中国
推荐引用方式 GB/T 7714	葛翔宇,丁建丽,王敬哲,等. 基于竞争适应重加权采样算法耦合机器学习的土壤含水量估算[J]. 新疆师范大学, 新疆大学,2018,38(10):1030001-1-1030001-8.
APA	葛翔宇,丁建丽,王敬哲,王飞,蔡亮红,&孙慧兰.(2018).基于竞争适应重加权采样算法耦合机器学习的土壤含水量估算.光学学报,38(10),1030001-1-1030001-8.
MLA	葛翔宇,et al."基于竞争适应重加权采样算法耦合机器学习的土壤含水量估算".光学学报 38.10(2018):1030001-1-1030001-8.