土壤有机质含量高光谱估测模型构建及精度对比

doi:10.12133/j.smartag.2020.2.3.201912-SA004

Smart Agriculture ›› 2020, Vol. 2 ›› Issue (3): 129-138.doi: 10.12133/j.smartag.2020.2.3.201912-SA004

土壤有机质含量高光谱估测模型构建及精度对比

刘恬琳¹(), 朱西存^1,²(), 白雪源¹, 彭玉凤¹, 李美炫¹, 田中宇¹, 姜远茂³, 杨贵军⁴

^1.山东农业大学资源与环境学院，山东泰安 271018
^2.土肥资源高效利用国家工程实验室，山东泰安 271018
^3.山东农业大学园艺科学与工程学院，山东泰安 271018
^4.国家农业信息化技术工程研究中心，北京 100097

收稿日期:2019-12-23 修回日期:2020-06-26 出版日期:2020-09-30
基金资助:
国家重点研发计划项目(2017YFE0122500);国家自然科学基金(41671346);山东省重大科技创新工程项目(2018CXGC0209);山东省泰山学者工程专项经费;山东省“双一流”资助项目(SYL2017XTTD02)
作者简介:刘恬琳（1995－），女，硕士，研究方向为农业定量遥感。E-mail：ltlstudy@163.com。
通信作者:

Hyperspectral Estimation Model Construction and Accuracy Comparison of Soil Organic Matter Content

LIU Tianlin¹(), ZHU Xicun^1,²(), BAI Xueyuan¹, PENG Yufeng¹, LI Meixuan¹, TIAN Zhongyu¹, JIANG Yuanmao³, YANG Guijun⁴

^1.College of Resources and Environment, Shandong Agricultural University, Tai'an 271018, China
^2.National Key Laboratory for Efficient Utilization of Soil and Fertilizer Resources, Tai'an 271018, China
^3.College of Horticulture Science and Engineering, Shandong Agricultural University, Tai'an 271018, China
^4.National Engineering Research Center for Information Technology in Agriculture, Beijing 100097, China

Received:2019-12-23 Revised:2020-06-26 Online:2020-09-30

摘要/Abstract

摘要：

土壤有机质含量对作物的生长发育有着显著影响。为实现对苹果果园土壤有机质含量快速、实时估测，本研究以山东省烟台市栖霞市苹果园为研究区，采集100个土壤样本，利用ASD FieldSpec3便携式地物光谱仪获取其高光谱反射率，利用定量化学方法测定土壤有机质含量。采用移动平均法对高光谱数据进行预处理，分析果园土壤的反射光谱特征，研究光谱反射率与其有机质含量的相关关系，筛选土壤有机质含量的敏感波长并构建光谱指数后，分别建立多元线性回归模型（MLR）、支持向量机（SVM）和随机森林（RF）模型，并对模型精度进行验证比较。结果表明，筛选出的土壤有机质含量的敏感波长为678、709、1931、1939、1996和2201 nm。用筛选出的波长构建光谱参数，最终构建的光谱指数分别为NDSI_{（678，709）}、NDSI_{（678，1931）}、NDSI_{（678，2201）}、NDSI_{（709，1939）}和NDSI_{（1939，2201）}。建立的MLR、SVM和RF回归模型中，以RF模型精度最优，其校正样本集R²为0.8804，RMSE为0.1423，RPD达到2.25；验证模型的R²为0.7466，RMSE为0.1266，RPD为1.79，建立的RF定量模型反演苹果果园土壤有机质含量效果较好。因此，可以利用RF方法快速预测苹果果园土壤有机质含量，了解土壤养分分布状况，指导农民合理施肥，从而提高果园生产管理效率。

关键词: 高光谱, 土壤有机质, 多元线性回归, 支持向量机, 随机森林, 模型

Abstract:

Soil organic matter (SOM) is an important source of crop growth, its content can reflect soil fertility status. In order to realize the fast and real-time estimation of the SOM, based on hyperspectral data, a rapid estimation model of SOM content in orchards was established. A total of 100 brown soil samples were collected from the apple orchard of Qixia county, Yantai city, Shandong province. After drying and grinding, the hyper-spectrum of the soil was measured in the laboratory using ASD FieldSpec. The spectral data was preprocessed by the method of moving average, and the spectral reflectance features of orchard soil were analyzed to study the correlation between spectral reflectance and its soil organic matter content. In order to enhance the correlation between relevant spectral parameters and soil indexes, the original data were processed by using the multivariate scattering correction, the first derivative and the first derivative of MSC. After the sensitive wavelengths of soil organic matter content were selected and the spectral indexes were constructed. Multiple linear regression models (MLR), support vector machines (SVM) and random forest (RF) models were respectively established. The estimation accuracy of the orchard soil organic matter estimation model was measured by the determination coefficient (R²), root mean square error (RMSE) and relative analysis error (RPD). The sensitive wavelengths of soil organic matter content selected were 678, 709, 1931, 1939, 1996 and 2201 nm. The spectral parameters were constructed using the selected wavelengths, which were NDSI_{(678, 709)}, NDSI_{(678, 1931)}, NDSI_{(678, 2201)}, NDSI_{(709, 1939)}, and NDSI_{(1939, 2201)}. These models established include MLR, SVM and RF model. The RF model had the best precision. The calibration sample R² was 0.8804, the RMSE was 0.1423 and RPD reached 2.25; the R²of the verification model was 0.7466, the RMSE was 0.1266, and the RPD was 1.79. The results showed that the fitting effect of the hyperspectral inversion model based on RF regression analysis was better than that based on MLR analysis and SVM regression analysis. As a promising and effective method, RF can play a vital role in predicting soil organic matter. The results can help understanding the distribution of soil nutrients, guiding farmers to apply fertilizer reasonably and improving the efficiency of orchard production and management.

Key words: hyperspectral, soil organic matter, multiple linear regression, support vector machine, random forest, model

中图分类号:

S127

刘恬琳, 朱西存, 白雪源, 彭玉凤, 李美炫, 田中宇, 姜远茂, 杨贵军. 土壤有机质含量高光谱估测模型构建及精度对比[J]. 智慧农业（中英文）, 2020, 2(3): 129-138.

LIU Tianlin, ZHU Xicun, BAI Xueyuan, PENG Yufeng, LI Meixuan, TIAN Zhongyu, JIANG Yuanmao, YANG Guijun. Hyperspectral Estimation Model Construction and Accuracy Comparison of Soil Organic Matter Content[J]. Smart Agriculture, 2020, 2(3): 129-138.

参考文献

1	DOTTO A C, DALMOLIN R S D, CATEN A T, et al. A systematic study on the application of scatter-corrective and spectral-derivative preprocessing for multivariate prediction of soil organic carbon by Vis-NIR spectra[J]. Geoderma, 2018, 314: 262-274.
2	冷疏影, 李秀彬. 土地质量指标体系国际研究的新进展[J]. 地理学报, 1999(2): 83-91.
2	LENG S, LI X. New progress of international research on land quality index system [J]. Journal of Geography, 1999(2): 83-91.
3	姚阔, 郭旭东, 周冬, 等. 高光谱遥感土地质量指标信息提取研究进展[J]. 地理与地理信息科学, 2014, 30(6): 7-12.
3	YAO K, GUO X, ZHOU D, et al. Advances in researches on hyperspectral remote sensing information extraction of land quality indicators[J]. Geography and Geo-Information Science, 2014, 30(6): 7-12.
4	傅伯杰. 我国生态系统研究的发展趋势与优先领域[J]. 地理研究, 2010, 29(3): 383-396.
4	FU B. Trends and priority areas in ecosystem research of China[J]. Geographical Research, 2010, 29(3): 383-396.
5	包青岭, 丁建丽, 王敬哲, 等. 基于随机森林算法的土壤有机质含量高光谱检测[J]. 干旱区地理, 2019, 42(6): 1404-1414.
5	BAO Q, DING J, WANG J, et al. Hyperspectral detection of soil organic matter content based on random forest algorithm[J]. Arid Land Geography, 2019, 42(6): 1404-1414.
6	李冠稳, 高小红, 肖能文, 等. 基于sCARS-RF算法的高光谱估算土壤有机质含量[J]. 发光学报, 2019, 40(8): 1030-1039.
6	LI G, GAO X, XIAO N, et al. Estimation soil organic matter contents with hyperspectra based on sCARS and RF algorithms[J]. Chinese Journal of Luminescence, 2019, 40(8): 1030-1039.
7	GEORGE K J, KUMAR S, RAJ R A. Soil organic carbon prediction using visible-near infrared reflectance spectroscopy employing artificial neural network modeling[J]. Current Science, 2020, 119(2): 377-381.
8	WEI L, YUAN Z, WANG Z, et al. Hyperspectral inversion of soil organic matter content based on a combined spectral index model[J]. Sensors, 2020, 20(10): 2777.
9	沈兰芝, 高懋芳, 闫敬文, 等. 基于SVR和PLSR的土壤有机质高光谱估测模型研究[J]. 中国农业信息, 2019, 31(1): 58-71.
9	SHEN L, GAO M, YAN J, et al. Estimation model of soil organic matter based on SVR and PLSR[J]. China Agricultural Informatics, 2019, 31(1): 58-71.
10	国佳欣, 赵小敏, 郭熙, 等. 基于PLSR-BP复合模型的红壤有机质含量反演研究[J]. 土壤学报, 2020, 57(3): 636-645.
10	GUO J, ZHAO X, GUO X, et al. Inversion of organic matter content in red soil based on PLSR-BP composite model[J]. Acta Pedologica Sinica, 2020, 57(3): 636-645.
11	张智韬, 劳聪聪, 王海峰, 等. 基于FOD和SVMDA-RF的土壤有机质含量高光谱预测[J]. 农业机械学报, 2020, 51(1): 156-167.
11	ZHANG Z, LAO C, WANG H, et al. Estimation of Desert Soil Organic Matter through hyperspectra based on Fractional-Order Derivatives and SVMDA-RF[J]. Transactions of the CSAM, 2020, 51(1): 156-167.
12	王延仓, 杨秀峰, 赵起超, 等. 二进制小波技术定量反演北方潮土土壤有机质含量[J]. 光谱学与光谱分析, 2019, 39(9): 2855-2861.
12	WANG Y, YANG X, ZHAO Q, et al. Quantitative inversion of soil organic matter content in Northern Alluvial soil based on binary wavelet transform[J]. Spectroscopy and Spectral Analysis, 2019, 39(9): 2855-2861.
13	王崇红, 陈冬生, 王燕. 烟台市苹果产业布局及结构优化建议[J]. 中国果树, 2019(6): 101-105.
13	WANG C, CHEN D, WANG Y. Suggestions on optimization of apple industry layout and structure in Yantai city [J]. China Fruits, 2019(6): 101-105
14	鲍士旦. 土壤农化分析[M]. 北京: 中国农业出版社, 1999.
14	BAO S. Soil agrochemical analysis [M]. Beijing: China Agricultural Press, 1999.
15	洪永胜, 于雷, 耿雷, 等. 应用DS算法消除室内几何测试条件对土壤高光谱数据波动性的影响[J]. 华中师范大学学报(自然科学版), 2016, 50(2): 303-308.
15	HONG Y, YU L, GENG L, et al. Using direct standardization algorithm to eliminate the effect of laboratory geometric parameters on soil hyperspectral data fluctuate characteristic[J]. Journal of Central China Normal University (Natural Sciences), 2016, 50(2): 303-308.
16	O'ROURKE S M, HOLDEN N M. Determination of soil organic matter and carbon fractions in forest top soils using spectral data acquired from visible-near infrared hyperspectral images[J]. Soil Science Society of America Journal, 2012, 76: 586-596.
17	周萍, 王润生, 阎柏琨, 等. 高光谱遥感土壤有机质信息提取研究[J]. 地理科学进展, 2008(5): 27-34.
17	ZHOU P, WANG R, YAN B, et al. Extraction of soil organic matter information by hyperspectral remote sensing[J]. Progress in Geography, 2008(5): 27-34.
18	洪永胜, 朱亚星, 苏学平, 等. 高光谱技术联合归一化光谱指数估算土壤有机质含量[J]. 光谱学与光谱分析, 2017, 37(11): 3537-3542.
18	HONG Y, ZHU Y, SU X, et al. Estimation of soil organic matter content using hyperspectral techniques combined with normalized difference spectral index[J]. Spectroscopy and Spectral Analysis, 2017, 37(11): 3537-3542.
19	崔霞, 宋清洁, 张瑶瑶, 等. 基于高光谱数据的高寒草地土壤有机碳预测模型研究[J]. 草业学报, 2017, 26(10): 20-29.
19	CUI X, SONG Q, ZHANG Y, et al. Estimation of soil organic carbon content in alpine grassland using hyperspectral data[J]. Acta Prataculturae Sinica, 2017, 26(10): 20-29.
20	于雷, 洪永胜, 周勇, 等. 高光谱估算土壤有机质含量的波长变量筛选方法[J]. 农业工程学报, 2016, 32(13): 95-102.
20	YU L, HONG Y, ZHOU Y, et al. Wavelength variable selection methods for estimation of soil organic matter content using hyperspectral technique[J]. Transactions of the CSAE, 2016, 32(13): 95-102.
21	韩兆迎, 朱西存, 刘庆, 等. 黄河三角洲土壤有机质含量的高光谱反演[J]. 植物营养与肥料学报, 2014, 20(6): 1545-1552.
21	HAN Z, ZHU X, LIU Q, et al. Hyperspectral inversion models for soil organic matter content in the Yellow River Delta[J]. Journal of Plant Nutrition and Fertilizers, 2014, 20(6): 1545-1552.
22	GAO L, ZHU X, HAN Z, et al. Spectroscopy-based soil organic matter estimation in brown forest soil areas of the Shandong Peninsula, China[J]. Pedosphere, 2019, 29(6): 810-818.
23	CAO S, ZHU X, LI C, et al. Estimating total nitrogen content in brown soil of orchard based on hyperspectrum[J]. Open Journal of Soil Science, 2017, 7: 203-215.

[1]	赵毓, 任艺平, 朴欣茹, 郑丹阳, 李东明. 基于改进ShuffleNet V2的轻量级防风药材道地性智能识别[J]. 智慧农业(中英文), 2023, 5(2): 104-114.
[2]	石杰锋, 黄为, 范协洋, 李修华, 卢阳旭, 蒋柱辉, 王泽平, 罗维, 张木清. 基于多种机器学习算法预测广西蔗区甘蔗产量[J]. 智慧农业(中英文), 2023, 5(2): 82-92.
[3]	魏永康, 杨天聪, 丁信尧, 高越之, 袁鑫茹, 贺利, 王永华, 段剑钊, 冯伟. 基于不同空间分辨率无人机多光谱遥感影像的小麦倒伏区域识别方法[J]. 智慧农业(中英文), 2023, 5(2): 56-67.
[4]	赖佳政, 李贝贝, 程翔, 孙丰, 陈炬廷, 王晶, 张芊, 叶协锋. 基于无人机高光谱遥感的烤烟叶片叶绿素含量估测[J]. 智慧农业(中英文), 2023, 5(2): 68-81.
[5]	夏雪, 柴秀娟, 张凝, 周硕, 孙琦鑫, 孙坦. 用于边缘计算设备的果树挂果量轻量化估测模型[J]. 智慧农业(中英文), 2023, 5(2): 1-12.
[6]	杨斌, 韩佳伟, 杨霖, 任青山, 杨信廷. 中国低碳冷链物流发展水平评价体系研究[J]. 智慧农业(中英文), 2023, 5(1): 44-51.
[7]	计洁, 金洲, 王儒敬, 刘海燕, 李志远. 基于递进式卷积网络的农业命名实体识别方法[J]. 智慧农业(中英文), 2023, 5(1): 122-131.
[8]	付虹雨, 王薇, 廖澳, 岳云开, 许明志, 王梓薇, 陈建福, 佘玮, 崔国贤. 基于无人机遥感表型监测的苎麻优质种质资源筛选方法[J]. 智慧农业(中英文), 2022, 4(4): 74-83.
[9]	许钰林, 康孟珍, 王秀娟, 华净, 王浩宇, 沈震. 基于深度学习的玉米和大豆期货价格智能预测[J]. 智慧农业(中英文), 2022, 4(4): 156-163.
[10]	范承志, 王梓文, 杨兴超, 罗永开, 徐学欣, 郭斌, 李振海. 基于地物高光谱和无人机多光谱的黄河三角洲土壤盐分机器学习反演模型[J]. 智慧农业(中英文), 2022, 4(4): 61-73.
[11]	夏烨, 雷哓晖, 祁雁楠, 徐陶, 袁全春, 潘健, 姜赛珂, 吕晓兰. 基于改进Ghost-YOLOv5s-BiFPN算法检测梨树花序[J]. 智慧农业(中英文), 2022, 4(3): 108-119.
[12]	庄家煜, 许世卫, 李杨, 熊露, 刘克宝, 钟志平. 基于深度学习的多种农产品供需预测模型[J]. 智慧农业(中英文), 2022, 4(2): 174-182.
[13]	张楷, 韩书庆, 程国栋, 吴赛赛, 刘继芳. 基于高斯混合-隐马尔科夫融合算法识别奶牛步态时相[J]. 智慧农业(中英文), 2022, 4(2): 53-63.
[14]	李少波, 杨玲, 于辉辉, 陈英义. 水下鱼类品种识别模型与实时识别系统[J]. 智慧农业(中英文), 2022, 4(1): 130-139.
[15]	赵晋陵, 杜世州, 黄林生. 联合多源多时相卫星影像和支持向量机的小麦白粉病监测方法[J]. 智慧农业(中英文), 2022, 4(1): 17-28.

土壤有机质含量高光谱估测模型构建及精度对比

Hyperspectral Estimation Model Construction and Accuracy Comparison of Soil Organic Matter Content

在线阅读

知网下载

本地下载

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价