基于高光谱小波能量特征向量估算湖滨绿洲表层土壤有机碳含量

doi:10.3964/j.issn.1000-0593(2023)12-3853-09

摘要
参考文献
相关文章 (3)

全文: PDF (8191 KB)
输出: BibTeX | EndNote (RIS)

摘要：高光谱分析能够高效的估算土壤有机碳含量，连续小波变换(CWT)，在高光谱数据的噪声去除和有效信息提取方面具有独特优势，但是经过连续小波变换后的光谱数据被分解为多个尺度，单一分解尺度信息不能代表不同分解尺度信息，如何充分利用多分解尺度的小波系数，成为高光谱估算土壤有机碳含量的难题。博斯腾湖是我国最大的内陆淡水湖，湖滨绿洲作为重要的水陆交错带，具有独特的空间结构和时间结构，在维持和恢复湖泊生态系统健康方面发挥着重要作用。以博斯腾湖湖滨绿洲为研究区，于2020年10月采集138份深度为0~20 cm表层土壤样本，剔除3个异常值样品，得到135个有效样品，室外采集土壤样本光谱，并通过重铬酸钾-外加热法测定土壤有机碳含量；将土壤样本的光谱反射率进行Savitzky-Golay平滑滤波处理，以Gaussian4为小波基函数进行连续小波变换，将土壤高光谱数据转换为10个分解尺度的小波系数。利用相关性分析法(CC)、稳定自适应重加权采样(sCARS)、竞争自适应重加权采样(CARS)、连续投影算法(SPA)、遗传算法(GA)等5种特种波段筛选方法进一步降低噪音，消除冗余信息，逐尺度计算小波系数的均方根作为小波能量特征(EF)，将10个尺度的小波能量特征组成小波能量特征向量(EF_V)，基于小波能量特征向量建立BP神经网络模型(BPNN)。结果表明，连续小波变换可以有效提高光谱反射率与土壤有机碳含量间的相关性，1~3分解尺度相关性较差，4~10分解尺度的相关性较好，相关系数平均值提升43.66%，相关系数最大值平均提升67.93%。CC算法筛选的特征波段主要分布于在400~1 500 nm可见光及近红外短波；sCARS、CARS算法筛选的特征波段集中于1 500~2 500 nm近红外长波；SPA算法筛选的特征波段集中于760~2 500 nm近红外波段；GA算法得到的特征波段基本均匀分布于400~2 500 nm。高光谱小波能量特征向量EF_V可以较好估算湖滨绿洲表层土壤有机碳含量，6种模型的训练集与验证集R²平均值分别为0.73、0.74，RMSE平均值分别为7.64、7.28，RPD平均值为1.95。模型精度表现为，CC-EF_V-BPNN＞sCARS-EF_V-BPNN＞Full-spectrum-EF_V-BPNN＞CARS-EF_V-BPNN＞GA-EF_V-BPNN＞SPA-EF_V-BPNN。连续小波变换结合特征变量筛选方法，提取小波能量特征向量EF_V，有效降低光谱数据维度与高光谱小波能量特征向量模型复杂度，对于快速估算表层土壤有机碳含量具有重要参考价值。

关键词：土壤有机碳含量；小波能量特征向量；分解尺度；特征波段筛选；湖滨绿洲

Abstract：Soil hyperspectral technique could estimate soil organic carbon content efficiently. Continuous wavelet transform had unique advantages in noise removal and effective information extraction of hyperspectral data. However, the spectral data after continuous wavelet transform was decomposed into multiple scales, and the information of a single decomposition scale could not represent the information of different decomposition scales. Making full use of the wavelet coefficients of multiple decomposition scales becomes a difficult problem for hyperspectral estimation of soil organic carbon content. Lake Bosten was the largest inland freshwater lake in China, and the lakeside oasis, as an important interlacing zone between land and water, had a unique spatial and temporal structure and played an important role in maintaining and restoring the health of the lake ecosystem. The study area was the lakeside oasis of Bosten Lake. 138 surface soil samples were collected in September 2020 at a depth of 0~20 cm, 3 outlier samples were excluded to obtain 135 useful samples, soil sample spectra were collected outdoors, and soil organic carbon content was determined by potassium dichromate-external heating method. The continuous wavelet transform was then performed with Gaussian4 as the wavelet basis function to convert the soil hyper spectrum into wavelet coefficients at 10 decomposition scales, and the correlation coefficient method, Stability Competitive Adaptive Reweighted Sampling, Competitive Adaptive Reweighted Sampling, Successive Projections Algorithm, Genetic Algorithm, five special wave filtering methods to further reduce noise and eliminate redundant information, calculate the root mean square of wavelet coefficients as wavelet energy feature scale by scale, and form a wavelet energy feature vector (Energy Feature Revector) with 10 scales of wavelet energy features, and build a BP neural network model (BP neural network model) based on the wavelet energy feature vector. The result showed that wavelet continuous transform could effectively improve the correlation between spectral reflectance and soil organic carbon content, with poor correlation at the 1~3 decomposition scale and good correlation at the 4~10 decomposition scale, with an average increase of 43.66% in the correlation coefficient and an average increase of 67.93% in the maximum value of the correlation coefficient. The feature band screening CC algorithm was mainly distributed in 400~1 500 nm visible and NIR short wavelength; sCARS and CARS algorithms were concentrated in 1 500~2 500 nm NIR long wavelength; SPA algorithm was concentrated in 760~2 500 nm NIR band; GA algorithm was uniformly distributed in 400~2 500 nm. The hyperspectral wavelet energy feature could better estimate the organic carbon content of the surface soil of the lakeshore oasis, and the mean R² values of the training and validation sets of the six models were 0.73 and 0.74, respectively; the mean RMSE values were 7.64 and 7.28, respectively; and the mean RPD value was 1.95. The model accuracy showed that CC-EF_V-BPNN＞sCARS-EF_V-BPNN＞Full-spectrum-EF_V-BPNN＞CARS-EF_V-BPNN＞GA-EF_V-BPNN＞SPA-EF_V-BPNN. The continuous wavelet transform combined with the feature variable screening method to extract the wavelet energy feature vector effectively reduces the spectral data dimensionality and hyperspectral wavelet energy feature model complexity, an important reference value for rapidly estimating surface soil organic carbon content.

Key words：Soil organic carbon content; Wavelet energy feature vector; Decomposition scale; Characteristic band screening; Lakeside Oasis

收稿日期: 2022-09-25 修订日期: 2023-03-29

中图分类号:

S151.9

基金资助: 新疆维吾尔自治区自然科学基金项目(2022D01A214)，国家自然科学基金项目(41661047)资助

通讯作者: 李新国 E-mail: onlinelxg@163.com

作者简介: 孟珊，1997年生，新疆师范大学地理科学与旅游学院硕士研究生 e-mail: mengshan1997@163.com

引用本文:

孟珊，李新国. 基于高光谱小波能量特征向量估算湖滨绿洲表层土壤有机碳含量[J]. 光谱学与光谱分析, 2023, 43(12): 3853-3861.
MENG Shan, LI Xin-guo. Estimation of Surface Soil Organic Carbon Content in Lakeside Oasis Based on Hyperspectral Wavelet Energy Feature Vector. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(12): 3853-3861.

链接本文:

https://www.gpxygpfx.com/CN/10.3964/j.issn.1000-0593(2023)12-3853-09 或 https://www.gpxygpfx.com/CN/Y2023/V43/I12/3853

[1] Greiner L, Keller A, Grêt-Regamey A, et al. Land Use Policy, 2017, 69: 224.
[2] Drobnik T, Greiner L, Keller A, et al. Ecological Indicators, 2018, 94: 151.
[3] LIU Xing-hua, ZHANG Hai-bo, LI Yuan, et al(刘兴华, 章海波, 李远, 等). Acta Pedologica Sinica(土壤学报), 2019, 56(2): 374.
[4] Chai Y J, Zeng X B, E Shengzhe, et al. Catena, 2019, 173: 312.
[5] YUAN Ya-ru, LI Na, ZOU Wen-xiu, et al(苑亚茹, 李娜, 邹文秀, 等). Acta Ecologica Sinica(生态学报), 2018, 38(17)：6025.
[6] Chen Kai, Li Chuang, Tang Rongnian. Industrial Crops and Products, 2017, 108: 831.
[7] ZHAO Ming-song, XIE Yi, LU Long-mei, et al(赵明松, 谢毅, 陆龙妹, 等). Acta Pedologica Sinica(土壤学报), 2021, 58(1): 42.
[8] YU Lei, HONG Yong-sheng, ZHOU Yong, et al(于雷, 洪永胜, 周勇, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2016, 32(13): 95.
[9] YE Hong-yun, XIONG Hei-gang, ZHANG Fang, et al(叶红云, 熊黑钢, 张芳, 等). Laser and Optoelectronics Progress(激光与光电子学进展) , 2019, 56(5): 115.
[10] Gu X H, Wang Y C, Sun Q, et al. Computers and Electronics in Agriculture, 2019, 167: 105053.
[11] Lin L X, Wang Y J, Teng J Y, et al. Environmental Monitoring and Assessment, 2016, 188(2): https://doi.org/10.1007/s10661-016-5107-8.
[12] Li R, Gao M, Xu Z X, et al. Hyper-Spectral Estimation of Soil Organic Matter in Apple Orchard Based on CWT, IOP Conference Series Earth and Environmental Science, 2021, 734: 012030.
[13] ZHAO Hai-long, GAN Shu, YUAN Xi-ping, et al(赵海龙, 甘淑, 袁希平, 等). Acta Optica Sinica, 2022, 42(22): 209.
[14] GOU Yu-xuan, ZHAO Yun-ze, LI Yong, et al(勾宇轩, 赵云泽, 李勇, 等). Transactions of the Chinese Society for Agricultural Machinery(农业机械学报), 2022, 53(3): 331.
[15] CAI Liang-hong, DING Jian-li(蔡亮红, 丁建丽). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2017, 33(16): 144.
[16] Wang L S, Wang R J, Lu C P, et al. Infrared Physics and Technology, 2019, 102: 103045.
[17] Banskota A, Falkowski M J, Smith A M S, et al. IEEE Transactions on Geoscience and Remote Sensing, 2016, 55(3): 1526.
[18] ZHANG Tao, YU Lei, YI Jun, et al(章涛, 于雷, 易军, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2019, 39(10): 3217.
[19] YE Qin, JIANG Xue-qin, LI Xi-can, et al(叶勤, 姜雪芹, 李西灿, 等). Transactions of the Chinese Society for Agricultural Machinery(农业机械学报), 2017, 48(3): 164.
[20] NIU Fang-peng, LI Xin-guo, Mamattursun·Eziz, et al(牛芳鹏, 李新国, 麦麦提吐尔逊·艾则孜, 等). Journal of Zhejiang University(Agriculture and Life Sciences)[浙江大学学报(农业与生命科学版)], 2021, 47(5): 673.
[21] XU Yong-mei, LIU Hua, WANG Xi-he(许咏梅, 刘骅, 王西和). Resources Science(资源科学), 2016, 38(7): 1246.
[22] LI Jin-pu, YU Xiu-bo, XIA Shao-xia, et al(李瑾璞, 于秀波, 夏少霞, 等). Acta Ecologica Sinica(生态学报), 2020, 40(24)：8928.
[23] ZHANG Zi-peng, DING Jian-li, WANG Jing-zhe(张子鹏, 丁建丽, 王敬哲). Acta Optica Sinica(光学学报), 2019, 39(2): 391.
[24] LI Guan-wen, GAO Xiao-hong, XIAO Neng-wen, et al(李冠稳, 高小红, 肖能文，等). Acta Optica Sinica(光学学报), 2019, 39(9): 361.