Please wait a minute...
 
自然资源遥感  2023, Vol. 35 Issue (3): 170-178    DOI: 10.6046/zrzyyg.2022376
  技术方法 本期目录 | 过刊浏览 | 高级检索 |
基于分数阶微分的土壤全氮高光谱反演
陈昊宇(), 项磊(), 高贺, 牟金燚, 索晓晶, 滑博伟
中国地质调查局呼和浩特自然资源综合调查中心,呼和浩特 010010
Hyperspectral inversion of total nitrogen content in soils based on fractional order differential
CHEN Haoyu(), XIANG Lei(), GAO He, MU Jinyi, SUO Xiaojing, HUA Bowei
Hohhot General Survey of Natural Resources Center, China Geological Survey, Hohhot 010010, China
全文: PDF(4478 KB)   HTML  
输出: BibTeX | EndNote (RIS)      
摘要 

为快速无损地获取托克托县土壤全氮含量,满足当今精准农业的要求,文章以研究区内120个采样点的土壤全氮含量与高光谱数据为数据源,利用分数阶微分(1~2阶)间隔0.1对光谱数据进行处理,筛选敏感波段,利用支持向量机(support vector machine,SVM)与BP神经网络模型共建立24个土壤全氮反演模型,结果表明: ①经过分数阶微分处理后,光谱的波峰、波谷处信息被放大,随分解尺度的增加,其余波段的反射率逐渐趋于0; ②原始光谱与土壤全氮的皮尔森相关系数r=0.61,经分数阶微分处理后,在1.1阶处达到最大值r=-0.67,绝对值较之前提升了0.06; ③BP神经网络预测模型结果优于SVM预测模型结果,本研究最佳土壤全氮预测模型为1.1阶微分处理后建立的BP神经网络模型,建模集R2为0.75,均方根误差(root mean square error,RMSE)为0.16,验证集R2为0.71,RMSE为0.16,相对分析误差(relative percent deviation,RPD)为2.06,可有效反演当地土壤全氮含量,相对于原始光谱建立的BP神经网络模型精度有较高提升。因此,利用1.1阶微分处理后的高光谱数据建立BP神经网络模型可实现对研究区土壤全氮含量的反演预测,可为当地精准农业的发展提供理论参考与技术支撑。
服务
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章
陈昊宇
项磊
高贺
牟金燚
索晓晶
滑博伟
关键词 ： 分数阶微分,  BP神经网络,  支持向量机,  精准农业    
Abstract：

This study aims to determine the total nitrogen content (TNC) in soils in Tuoketuo County quickly and nondestructively, thus meeting the requirements of precision agriculture. With the soil TNC and hyperspectral data of 120 sampling sites in the study area as the data source, this study processed the hyperspectral data using the 1~2 orders fractional order differential (FOD) interval of 0.1 to screen the sensitive wavebands. Then, this study built 24 inversion models for soil TNC using the support vector machine (SVM) and the back propagation neural network (BPNN). The results are as follows: ① After FOD processing, the information at the wave crests and troughs of the spectra was amplified, and the reflectance of the remaining wavebands approached zero gradually with an increase in the decomposition scale. ② The Pearson correlation coefficient between original spectra and soil TNC was r = 0.61. This correlation coefficient was up to a maximum of 0.67 at 1.1- order after FOD processing, increasing by 0.06. ③ The BPNN prediction models outperformed the SVM prediction models. The optimal soil TNC prediction model was the BPNN model built after 1.1-order differential processing. This model yielded an R2 of 0.75 and a root mean square error (RMSE) of 0.16 for the modeling set and an R2 of 0.71 and an RMSE of 0.16 for the verification set, with a relative percent deviation (RPD) of 2.06. This model produced effective inversion results of the soil TNC in the study area, with a much higher accuracy than the BPNN model built using original spectra. Therefore, the BPNN model built using hyperspectral data through 1.1-order differential processing allows for the inversion-based prediction of soil TNC in the study area, providing a theoretical reference and technical support for local precision agriculture.
Key words： fractional order differential    BPNN    SVM    precision agriculture
收稿日期: 2022-09-19      出版日期: 2023-09-19
ZTFLH:  TP79  
通讯作者: 项 磊(1983-),男,学士,高级工程师,主要从事自然资源调查、地质矿产研究。Email: xiang6280@163.com。
作者简介: 陈昊宇(1995-),男,硕士,主要从事自然资源调查、高光谱遥感研究。Email: chenhaoyu0807@163.com。
引用本文:   
陈昊宇, 项磊, 高贺, 牟金燚, 索晓晶, 滑博伟. 基于分数阶微分的土壤全氮高光谱反演[J]. 自然资源遥感, 2023, 35(3): 170-178.
CHEN Haoyu, XIANG Lei, GAO He, MU Jinyi, SUO Xiaojing, HUA Bowei. Hyperspectral inversion of total nitrogen content in soils based on fractional order differential. Remote Sensing for Natural Resources, 2023, 35(3): 170-178.
链接本文:  
https://www.gtzyyg.com/CN/10.6046/zrzyyg.2022376      或      https://www.gtzyyg.com/CN/Y2023/V35/I3/170
Fig.1  托克托县研究区概况
Fig.2  土壤采集分布示意图
类型 最大值/
(g·
kg-1)		 最小值/
(g·
kg-1)		 均值/
(g·
kg-1)		 中位数/
(g·
kg-1)		 标准
差		 偏度 峰度 变异
系数/
%
全氮 1.74 0.11 0.53 0.41 0.33 1.27 1.16 0.62
Tab.1  土壤养分描述性统计
Fig.3  光谱室内测量示意图
Fig.4  光谱反射率曲线
Fig.5  不同全氮含量光谱反射率
Fig.6  光谱对照图
Fig.7  相关性分布
阶数① 波长范围/nm 波段数
0 815~819,830~835 11
1 828~829,534~535,548~554 11
1.1 535~539,541~542,817~818,834~835,839 12
1.2 535~536,816~819,830~835,837,839~840 15
1.3 785,810,815~819,822,830~835,837,839~840 17
1.4 535~536,816~819,830~835 12
1.5 624,639,640,746~747,769,816~818 9
1.6 621,624~625,638~640,680,727,746~747,816~818 13
1.7 604,621,624~625,638~639,666,727,816~817 10
1.8 604,621,624,638~639 5
1.9 604,1 411~1 412,1 415 4
2 1 404,1 407~1 408 3
Tab.2  敏感波段筛选
数据集 最大值/
(g·kg-1)		 最小值/
(g·kg-1)		 均值/
(g·kg-1)		 标准差/
(g·kg-1)		 土样
数/个
建模集 1.74 0.12 0.52 0.29 90
验证集 1.44 0.11 0.52 0.36 30
Tab.3  全氮建模集和验证集描述性统计
阶数① 建模集 验证集 RPD
R2 RMSE R2 RMSE
0 0.28 0.22 0.27 0.28 1.18
1 0.44 0.19 0.33 0.26 1.27
1.1 0.55 0.16 0.42 0.22 1.50
1.2 0.54 0.17 0.41 0.23 1.43
1.3 0.45 0.19 0.35 0.26 1.27
1.4 0.52 0.17 0.38 0.24 1.38
1.5 0.43 0.19 0.38 0.25 1.32
1.6 0.50 0.17 0.34 0.26 1.27
1.7 0.47 0.18 0.32 0.26 1.27
1.8 0.39 0.28 0.23 0.28 1.18
1.9 0.28 0.22 0.13 0.31 1.06
2 0.28 0.21 0.15 0.30 1.10
Tab.4  SVM预测模型
阶数① 建模集 验证集 RPD
R2 RMSE R2 RMSE
0 0.58 0.25 0.58 0.31 1.06
1 0.66 0.23 0.64 0.22 1.50
1.1 0.75 0.16 0.71 0.16 2.06
1.2 0.74 0.19 0.65 0.23 1.43
1.3 0.73 0.19 0.56 0.21 1.57
1.4 0.75 0.45 0.51 0.31 1.06
1.5 0.69 0.22 0.58 0.3 1.10
1.6 0.75 0.20 0.53 0.31 1.06
1.7 0.69 0.47 0.63 0.28 1.18
1.8 0.60 0.25 0.63 0.29 1.14
1.9 0.54 0.26 0.37 0.33 1.00
2 0.53 0.26 0.46 0.32 1.03
Tab.5  BP神经网络预测模型
[1] 王炜超, 杨玮, 崔玉露, 等. 基于CatBoost算法与图谱特征融合的土壤全氮含量预测[J]. 农业机械学报, 2021, 52(s1):316-322.
Wang W C, Yang W, Cui Y L, et al. Prediction of soil total nitrogen based on CatBoost algorithm and fusion of image spectral features[J]. Transactions of the Chinese Society of Agricultural Machinery, 52(s1):316-322.
[2] 殷彩云, 白子金, 罗德芳, 等. 基于高光谱数据的土壤全氮含量估测模型对比研究[J]. 中国土壤与肥料, 2022(1):9-15.
Yin C Y, Bai Z J, Luo D F, et al. Comparative study on estimation models of soil total nitrogen content based on hyperspectral data[J]. Soil and Fertilizer Sciences in China, 2022(1):9-15.
[3] 钟浩, 李西灿, 翟浩然, 等. 基于表层土壤光谱的耕层土壤有机质间接估测[J]. 安徽农业大学学报, 2020, 47(3):421-426.
Zhong H, Li X C, Zhai H R, et al. Indirect estimation of organic matter content in plough layer based on topsoil spectrum[J]. Journal of Anhui Agricultural University, 2020, 47(3):421-426.
[4] 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.
Wang Y C, Yang X F, Zhao Q C, et al. Quantitative inversion of soil organic matter content in northern tidal soil by binary wavelet technique[J]. Spectroscopy and Spectral Analysis, 2019, 39(9):2855-2861.
doi: 10.3964/j.issn.1000-0593(2019)09-2855-07
[5] Dalai R C, Henry R J. Simultaneous determination of moisture,organic carbon,and totalnitrogen by in France spectrometry[J]. Soil Science Society of America Journal, 1986, 50:120-123.
doi: 10.2136/sssaj1986.03615995005000010023x
[6] Hummel J W, Sudduth K A, Hollinger S E. Soil moisture and organic matter prediction of surface and subsurface soils using an NIR soil sensor[J]. Computers and Electronics in Agriculture, 2001, 32(2):149-165.
doi: 10.1016/S0168-1699(01)00163-6
[7] Reeves J B, Mccarty G W, Meisinger J J. Near infrared reflectance spectroscopy for the analysis of agricultural soil[J]. Journal of Near Infrared Spectroscopy, 1997, 7(3):179-193.
doi: 10.1255/jnirs.248
[8] 卢艳丽, 白由路, 王磊, 等. 黑土土壤中全氮含量的高光谱预测分析[J]. 农业工程学报, 2010, 26(1):256-261.
Lu Y L, Bai Y L, Wang L, et al. Hyperspectral prediction analysis of total nitrogen content in black soil[J]. Transactions of the Chinese Society of Agricultural Engineering, 2010, 26(1):256-261.
[9] 张娟娟, 田永超, 姚霞, 等. 基于高光谱的土壤全氮含量估测[J]. 自然资源学报, 2011, 26(5):881-890.
Zhang J J, Tian Y C, Yao X, et al. Estimating soil total nitrogen content based on hyperspectral analysis technology[J]. Journal of Natural Resources, 2011, 26(5):881-890.
doi: 10.11849/zrzyxb.2011.05.015
[10] 赵燕东, 皮婷婷. 北京地区粘壤土全氮含量的光谱预测模型[J]. 农业机械学报, 2016, 47(3):144-149.
Zhao Y D, Pi T T. Spectral prediction model of total nitrogen content of clay loam in Beijing[J]. Trasactions of the Chinese Society of Agricaltrual Machinery, 2016, 47(3):144-149.
[11] 王世东, 石朴杰, 张合兵, 等. 基于高光谱的矿区复垦农田土壤全氮含量反演[J]. 生态学杂志, 2019, 38(1):294-301.
Wang S D, Shi P J, Zhang H B, et al. Inversion of soil total nitrogen content in reclaimed farmland of mining area based on hyperspectral imaging[J]. Chinese Journal of Ecology, 2019, 38(1):294-301.
[12] 赵启东, 葛翔宇, 丁建丽, 等. 结合分数阶微分技术与机器学习算法的土壤有机碳含量光谱估测[J]. 激光与光电子学进展, 2020, 57(15):253-261.
Zhao Q D, Ge X Y, Ding J L, et al. Combination fractional order differential and machine learning algorithm for spectral estimation of soil organic carbon content[J]. Laser and Optoelectronics Progress, 2020, 57(15):253-261.
[13] 李武耀, 买买提·沙吾提, 买合木提·巴拉提. 基于分数阶微分的土壤有机质含量高光谱反演研究[J]. 激光与光电子学进展, 2023, 60(7):404-411.
Li W Y, Mamat S, Maihemuti B. Hyperspectral inversion of soil organic matter content based on fractional differential[J]. Laser and Optoelectronics Progress, 2023, 60(7):404-411.
[14] 竞霞, 张腾, 邹琴, 等. 基于分数阶微分光谱指数的小麦条锈病遥感监测模型构建[J]. 农业工程学报, 2021, 37(17):142-151.
Jing X, Zhang T, Zou Q, et al. Construction of remote sensing monitoring model of wheat stripe rust based on fractional differential spectral index[J]. Transactions of the Chinese Society of Agricultural Engineering, 2021, 37(17):142-151.
[15] 赵慧, 李新国, 靳万贵, 等. 基于分数阶微分的博斯腾湖湖滨绿洲土壤电导率高光谱估算[J]. 甘肃农业大学学报, 2021, 56(1):118-125.
Zhao H, Li X G, Jin W G, et al. Hyperspectral estimation of soil conductivity in oasis along Bosten Lake based on fractional differential[J]. Journal of Gansu Agricultural University, 2021, 56(1):118-125.
[16] 田安红, 赵俊三, 张顺吉, 等. 基于分数阶微分的盐渍土电导率高光谱估算研究[J]. 中国生态农业学报(中英文), 2020, 28(4):599-607.
Tian A H, Zhao J S, Zhang S J, et al. Hyperspectral estimation of electrical conductivity of saline soil based on fractional differentiation[J]. Chinese Journal of Eco-Agriculture, 2020, 28(4):599-607.
[17] Wang X P, Zhang F, Kung H T, et al. New methods for improving the remote sensing estimation of soil organic matter content (SOMC) in the Ebinur Lake Wetland National Nature Reserve (ELWNNR) in northwest China[J]. Remote Sensing of Environment, 2018, 218:104-118.
doi: 10.1016/j.rse.2018.09.020
[18] 卢志宏, 刘辛瑶, 常书娟, 等. 基于BP神经网络的草原矿区表层土壤N/P高光谱反演模型[J]. 草业科学, 2018, 35(9):2127-2136.
Lu Z H, Liu X Y, Chang S J, et al. Hyperspectral inversion of the surface soil N/P rationin a grassland mining area based on the BP network[J]. Pratacultural Science, 2018, 35(9):2127-2136.
[19] 谭念, 孙一丹, 王学顺, 等. 基于主成分分析和支持向量机的木材近红外光谱树种识别研究[J]. 光谱学与光谱分析, 2017, 37(11): 3370-3374.
Tan N, Sun Y D, Wang X S, et al. Research on near infrared spectrum with principal component analysis and support vector machine for timber identification[J]. Spectroscopy and Spectral Analysis, 2017, 37(11): 3370-3374.
[20] 朱文静, 毛罕平, 周莹, 等. 基于高光谱图像技术的番茄叶片氮素营养诊断[J]. 江苏大学学报(自然科学版), 2014, 35(3): 290-294.
Zhu W J, Mao H P, Zhou Y, et al. Nitrogen nutrition diagnosis of tomato leaves based on hyperspectral image technology[J]. Journal of Jiangsu University (Natural Science Edition), 2014, 35(3): 290-294.
[21] 董毅, 程伟, 张燕平, 等. 基于SVM的先分类再回归方法及其在产量预测中的应用[J]. 计算机应用, 2010, 30(9):2310-2313.
Dong Y, Cheng W, Zhang Y P, et al. The method of classification and regression based on SVM and its application in yield prediction[J]. Journal of Computer Applications, 2010, 30(9):2310-2313.
doi: 10.3724/SP.J.1087.2010.02310
[22] 郭云开, 刘宁, 刘磊, 等. 土壤Cu含量高光谱反演的BP神经网络模型[J]. 测绘科学, 2018, 43(1):135-139,152.
Guo Y K, Liu N, Liu L, et al. Hyper-spectral inversion of soil Cu content based on BP neural network model[J]. Science of Surveying and Mapping, 2018, 43(1):135-139,152.
[23] Pan H, Wang X Y, Chen Q, et al. Application of BP neural network based on genetic algorithm[J]. Computer Applications, 2005, 25(12):2777-2779.
[24] 刘润, 张绍良, 侯湖平, 等. 基于思维进化优化BP神经网络的大豆叶片叶绿素含量高光谱反演[J]. 江苏农业科学, 2018, 46(13):212-216.
Liu R, Zhang S L, Hou H P, et al. Hyperspectral retrieval of chlorophyll content in soybean leaves based on mind evolutionary optimization BP neural network[J]. Jiangsu Agricultural Sciences, 2018, 46(13):212-216.
[25] 陈昊宇, 杨光, 韩雪莹, 等. 基于连续小波变换的土壤有机质含量高光谱反演[J]. 中国农业科技导报, 2021, 23(5):132-142.
Chen H Y, Yang G, Han X Y, et al. Hyperspectral inversion of soil organic matter content based on continuous wavelet transform[J]. China Agricultural Science and Technology Guide, 2021, 23(5):132-142.
[26] 刘杨, 冯海宽, 孙乾, 等. 基于无人机高光谱分数阶微分的马铃薯地上生物量估算[J]. 农业机械学报, 2020, 51(12):202-211.
Liu Y, Feng H K, Sun Q, et al. Estimation of potato above-groud biomass based on fractional differential of UAV hyperspectral[J]. Transactions of the Chinese Society of Agricultural Machinery, 2020, 51(12):202-211.
[27] 徐彬彬. 土壤剖面的反射光谱研究[J]. 土壤, 2000(6): 281-287.
Xu B B. Study on the reflection spectrum of soil profile[J]. Soil, 2000(6):281-287.
[28] 刘凡, 马玲, 杨光, 等. 灰漠土土壤全氮含量的高光谱特征分析及估测[J]. 新疆农业科学, 2017, 54(1):140-147.
Liu F, Ma L, Yang G, et al. Hyperspectral characteristic analysis and estimation of total nitrogen content in grey desert soil[J]. Xinjiang Agricultural Sciences, 2017, 54(1):140-147.
[29] 邵丽冰, 陈奕云, 徐璐, 等. 基于分数阶微分的土壤含水量高光谱响应特征与估测模型构建[J]. 测绘地理信息, 2022, 47(s1):131-136.
Shao L B, Chen Y Y, Xu L, et al. Analysis on soil moisture content hyperspectral response and construction of estimation model based on fractional-order derivative[J]. Surveying and Mapping Geographic Information, 2022, 47(s1):131-136.
[30] 李长春, 施锦锦, 马春艳, 等. 基于小波变换和分数阶微分的冬小麦叶绿素含量估算[J]. 农业机械学报, 2021, 52(8):172-182.
Li C C, Shi J J, Ma C Y, et al. Estimation of chlorophyll content in winter wheat based on wavelet transform and fractional differential[J]. Transactions of the Chinese Society of Agricultural Machinery, 2021, 52(8):172-182.
[1] 李志新, 王梦飞, 贾伟洁, 纪松, 王宇飞. 一种结合阴影信息的建筑物层数识别方法[J]. 自然资源遥感, 2023, 35(3): 97-106.
[2] 赵海岚, 蒙继华, 纪云鹏. 遥感技术在苹果园精准种植管理中的应用现状及展望[J]. 自然资源遥感, 2023, 35(2): 1-15.
[3] 赵连杰, 吴孟泉, 郑龙啸, 栾绍鹏, 赵贤峰, 薛明月, 刘佳燕, 刘晨曦. 胶东半岛北部海岸线时空变迁及驱动分析[J]. 自然资源遥感, 2022, 34(4): 87-96.
[4] 王春霞, 张俊, 李屹旭, Phoumilay. 复杂环境下GF-2影像水体指数的构建及验证[J]. 自然资源遥感, 2022, 34(3): 50-58.
[5] 邓静雯, 田义超, 张强, 陶进, 张亚丽, 黄升光. 机载LiDAR在红树林林分平均高估算中的应用[J]. 自然资源遥感, 2022, 34(3): 129-137.
[6] 朱琦, 郭华东, 张露, 梁栋, 刘栩婷, 万祥星. 基于多时相Landsat8影像的海南岛热带天然林类型遥感分类[J]. 自然资源遥感, 2022, 34(2): 215-223.
[7] 孙肖, 彭军还, 赵锋, 王晓阳, 吕洁, 张登峰. 基于空间统计学的高光谱遥感影像主成分选择方法[J]. 自然资源遥感, 2022, 34(2): 37-46.
[8] 赵晓伟, 黄杨, 汪永强, 储鼎. 基于无人机多光谱数据的玉米苗株估算[J]. 自然资源遥感, 2022, 34(1): 106-114.
[9] 冯东东, 张志华, 石浩月. 基于多元数据的省会城市城中村精细提取[J]. 自然资源遥感, 2021, 33(3): 272-278.
[10] 栗旭升, 刘玉锋, 陈冬花, 刘赛赛, 李虎. 结合图像特征的支持向量机高分一号云检测[J]. 国土资源遥感, 2020, 32(3): 55-62.
[11] 高凯旋, 焦海明, 王新闯. 融合影像纹理、光谱与地形特征的森林冠顶高反演模型[J]. 国土资源遥感, 2020, 32(3): 63-70.
[12] 刘义志, 赖华荣, 张丁旺, 刘飞鹏, 蒋小蕾, 曹庆安. 多特征混合核SVM模型的遥感影像变化检测[J]. 国土资源遥感, 2019, 31(1): 16-21.
[13] 徐念旭, 田庆久, 申怀飞, 徐凯健. 基于微分变换的高光谱马尾松和杉木识别[J]. 国土资源遥感, 2018, 30(4): 28-32.
[14] 张康, 黑保琴, 李盛阳, 邵雨阳. 基于CNN模型的遥感图像复杂场景分类[J]. 国土资源遥感, 2018, 30(4): 49-55.
[15] 魏英娟, 郑雄伟, 雷兵, 甘宇航. 基于样本模型的高分一号遥感影像云雾自动提取[J]. 国土资源遥感, 2017, 29(s1): 39-45.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
京ICP备05055290号-2
版权所有 © 2015 《自然资源遥感》编辑部
地址:北京学院路31号中国国土资源航空物探遥感中心 邮编：100083
电话:010-62060291/62060292 E-mail:zrzyyg@163.com
本系统由北京玛格泰克科技发展有限公司设计开发