Hyperspectral Estimation of Soil Organic Matter Content Based on Continuous Wavelet Transformation
Yumiti Maiming1, WANG Xue-mei1, 2*
1. College of Geographic Science and Tourism, Xinjiang Normal University, Urumqi 830054, China
2. Xinjiang Uygur Autonomous Region Key Laboratory “Xinjiang Arid Lake Environment and Resources Laboratory”, Urumqi
830054, China
Abstract：The hyperspectral estimation of soil organic matter content can quickly and accurately monitor soil fertility and provide a scientific basis for proper fertilization of modern agricultural production. Taking the cultivated soil in the delta oasis of Weigan-Kuqa river, Xinjiang as the research object, the original spectral reflectance (R) of the collected 98 soil samples was subjected to the traditional reciprocal logarithm lg(1/R), the first-order differential (R′) and the reciprocal logarithm first-order differential [lg(1/R)]′ mathematical transformations, and continuous wavelet transformation (Continuous Wavelet Transformation, CWT) processing based on Bior1.3 as the wavelet mother function through different scale decomposition. Correlation analysis was conducted between the treatment results and the measured soil organic matter content to screen out the characteristic bands and wavelet coefficients closely related to soil organic matter content under various transformations (p<0.01). With the original spectral reflectance, the characteristic band reflectance and the sensitive wavelet coefficient under different transformation treatments as independent variables and the soil organic matter content as dependent variables, partial least squares regression and support vector machine regression was used to estimate models of soil organic matter content. The results showed that: (1) Various spectral transformation methods can effectively improve the sensitivity between the spectrum and the content of soil organic matter. The correlation between the soil spectral reflectance and the organic matter content after continuous wavelet transformation has been significantly improved, and the correlation coefficient has been increased from 0.39 to 0.54 (p<0.01). (2) The support vector machine regression model built by the traditional [lg(1/R)]′ transformation has a higher coefficient of determination (R2) than the model built by lg(1/R) and R′ transformation, showing the reciprocal logarithm first-order differential transformation can help improve the accuracy of the estimation model, and the accuracy and stability of the support vector machine regression model were higher than that of the partial least squares regression model. (3) After continuous wavelet transformation decomposition, the estimation accuracy and stability of the models were obviously improved by using the sensitive wavelet coefficients of the original spectral reflectance at different scales as independent variables. The decision coefficient (R2), root mean square error (RMSE) and relative analysis error (RPD) of the CWT-23-SVMR model were 0.84, 1.48 and 2.11 respectively. The model has the highest accuracy and excellent predictive ability. After multiple transformation processing, hyperspectral data can effectively remove white noise. In contrast, continuous wavelet transformation processing is more suitable for mining effective soil information than the traditional mathematical transformation method, to realize the effective separation of approximate features and detailed features of spectral signals, and the established inversion model can more accurately estimate soil organic matter content.
Key words：Continuous wavelet transformation; Hyperspectral estimation; Decomposition scale; Soil organic matter; The delta oasis of Weigan-Kuqa rivers
[1] YANG Yue-chao, ZHAO Ying-jun, QIN Kai, et al(杨越超, 赵英俊, 秦 凯, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2019, 35(20): 94.
[2] 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.
[3] ZHOU Qian-qian, DING Jian-li, TANG Meng-ying, et al(周倩倩, 丁建丽, 唐梦迎, 等). Acta Pedologica Sinica(土壤学报), 2018, 55(2): 313.
[4] ZHANG Dong-hui, ZHAO Ying-jun, QIN Kai, et al(张东辉, 赵英俊, 秦 凯, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2018, 34(20): 141.
[5] NAN Feng, ZHU Hong-fen, BI Ru-tian（南 锋，朱洪芬，毕如田）. Scientia Agricultura Sinica(中国农业科学), 2016, 49(11): 2126.
[6] Shen L, Gao M, Yan J, et al. Remote Sensing, 2020, 12(7): 1206.
[7] ZHANG Rui, LI Zhao-fu, PAN Jian-jun(张 锐, 李兆富, 潘剑君). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2017, 33(1): 175.
[8] 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.
[9] YE Hong-yun, XIONG Hei-gang, ZHANG Fang, et al(叶红云, 熊黑钢, 张 芳, 等). Advances in Laser & Optoelectronics Progress(激光与光电子学进展), 2019, 56(5): 115.
[10] WANG Yong-min, LI Xi-can , TIAN Lin-ya, et al(王永敏, 李西灿, 田林亚, 等). Remote Sensing for Land and Resources(国土资源遥感), 2019, 31(1): 110.
[11] DENG Kai, DING Jian-li, YANG Ai-xia, et al(邓 凯, 丁建丽, 杨爱霞, 等). Journal of Ecology(生态学报), 2016, 36(20): 6387.
[12] 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.
[13] Asim Biswas. Catena, 2018, 160: 24.
[14] Shen L, Gao M, Yan J, et al. Remote Sensing, 2020, 12(7): 1206.
[15] Qiao X X, Wang C, Feng M C, et al. Spectroscopy Letters, 2017, 50(3): 156.
[16] Li R, Gao M, Zexin X U, et al. Hyperspectral Estimation of Soil Organic Matter in Apple Orchard Based on CWT. IOP Conference Series: Earth and Environmental Science. IOP Publishing, 2021, 734(1): 012030.
