基于随机蛙跳波段选择算法的土壤铅含量高光谱估测

doi:10.3964/j.issn.1000-0593(2023)10-3302-08

摘要
参考文献
相关文章 (15)

全文: PDF (4253 KB)
输出: BibTeX | EndNote (RIS)

摘要：高光谱数据中存在的大量冗余信息对高光谱估测精度产生较大影响。旨在寻求特征波段筛选的最佳算法，以实现土壤重金属铅含量的准确监测，为土壤污染防治提供参考。以新疆渭干河-库车河三角洲绿洲土壤重金属铅含量与光谱数据为数据源，利用蒙特卡洛交叉验证（MCCV）算法确定92个有效土壤样品，通过相关分析选取倒数对数一阶微分变换处理的光谱数据，采用随机蛙跳（RF）算法，并结合竞争性自适应重加权（CARS）算法、迭代保留有效信息变量（IRIV）算法及连续投影算法（SPA），构建RF-CARS、RF-IRIV及RF-SPA三种算法对波段进行筛选。以倒数对数一阶微分变换处理下的特征波段反射率为自变量，土壤重金属铅含量为因变量，采用极端梯度提升（XGBoost）和地理加权回归（GWR）方法构建土壤重金属铅含量估测模型。结果表明：（1）光谱变换处理可有效增强光谱与土壤铅含量的敏感性，其中倒数对数一阶微分变换后的土壤光谱特征更为明显，相关系数可达到0.620（p＜0.001）。（2）RF-CARS、RF-IRIV及RF-SPA算法分别从高光谱数据中筛选出6、9和7个特征波段，全部位于近红外光谱区域，3种算法具有较强的特征提取能力，极大减少光谱数据中的冗余信息。（3）基于RF-IRIV算法构建的土壤铅含量估测模型的精度和稳定性高于RF-CARS和RF-SPA算法构建的模型，说明RF-IRIV算法能更为准确的保留与土壤铅含量相关的波段。此外，GWR模型的性能优于XGBoost模型，构建的RF-IRIV-GWR模型具有较好的预测能力，可作为研究区土壤铅含量的最优估测模型，其验证集的决定系数（R²）为0.892，均方根误差（RMSE）为0.825 mg·kg^-1，相对分析误差（RPD）为3.09。基于随机蛙跳（RF）与迭代保留有效信息变量（IRIV）算法，结合地理加权回归（GWR）建模方法在快速准确估测土壤铅含量方面具有一定优势，可进行土壤重金属污染的动态监测。

关键词：特征波段；随机蛙跳算法；竞争性自适应重加权算法；迭代保留有效信息变量算法；连续投影算法；极端梯度提升；地理加权回归；土壤铅

Abstract：Due to the large amount of redundant information in hyperspectral data, it greatly impacts the accuracy of hyperspectral estimation. The purpose of this study is to find the best algorithm for the screening of feature bands to realize the accurate monitoring of the lead content of heavy metals in soil and to provide a reference for soil pollution prevention and control. The lead contents and spectral data in the oasis soils of the Weigan-Kuqa river delta in Xinjiang were used as data sources, and 92 valid soil samples were identified using the Monte Carlo cross-validation algorithm (MCCV), and the spectral data processed by the first-order differential transformation of the reciprocal logarithm are selected through correlation analysis. The random frog (RF) algorithm is combined with the competitive adaptive reweighted sampling (CARS) algorithm. The iteratively retains informative variables (IRIV) algorithm and the successive projections algorithm (SPA). The RF-CARS, RF-IRIV and RF-SPA algorithms are constructed to screen the bands. Taking the reflectivity of feature bands as the independent variable and the content of heavy metal lead in the soil as the dependent variable, the extreme gradient boosting (XG Boost) and geographically weighted regression (GWR) methods were used to construct the estimation model of the lead content in the soil. The results show that: (1) The spectral transformation treatment can effectively enhance the sensitivity of the spectrum and lead content. The spectral characteristics after the first-order differential transformation of the reciprocal logarithm are obvious, and the correlation coefficient can reach 0.620 (p＜0.001). (2) RF-CARS, RF-IRIV and RF-SPA algorithms extract 6, 9 and 7 feature bands from hyperspectral data, all located in the near-infrared spectral region. The three algorithms have strong feature extraction ability, greatly reducing redundant information in spectral data. (3) The accuracy and stability of the soil lead content estimation model constructed based on that the RF-IRIV algorithm are higher than those constructed by RF-CARS and RF-SPA, showing the RF-IRIV algorithm can more accurately retain the bands related to soil lead content. In addition, the performance of the GWR model is better than that of the XGBoost model, and the constructed RF-IRIV-GWR model has the good predictive ability, which can be used as the optimal estimation model for soil lead content in the study area. The R², RMSE and RPD of the validation set of the RF-IRIV-GWR model are 0.892, 0.825 mg·kg^-1 and 3.09 respectively. Based on the random frog (RF) and iteratively retains informative variables (IRIV) algorithm combined with geographically weighted regression (GWR) modeling method, it has certain advantages in quickly and accurately estimating soil lead content, which can be used for dynamic monitoring of soil heavy metal pollution.

Key words：Feature band; Random frog algorithm; Competitive adaptive reweighted sampling algorithm; Iteratively retaining informative variables algorithm; Successive projections algorithm; Extreme gradient boosting; Geographically weighted regression; Soil lead

收稿日期: 2022-05-10 修订日期: 2022-10-31

中图分类号:

X53

基金资助: 国家自然科学基金项目（41561051），新疆维吾尔自治区自然科学基金项目（2020D01A79）资助

通讯作者: 王雪梅 E-mail: wangxm_1225@sina.com

作者简介: 安柏耸，女，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: aaanbs@163.com

引用本文:

安柏耸，王雪梅，黄晓宇，卡吾恰提·白山. 基于随机蛙跳波段选择算法的土壤铅含量高光谱估测[J]. 光谱学与光谱分析, 2023, 43(10): 3302-3309.
AN Bai-song, WANG Xue-mei, HUANG Xiao-yu, KAWUQIATI Bai-shan. Hyperspectral Estimation of Soil Lead Content Based on Random Frog Band Selection Algorithm. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(10): 3302-3309.

链接本文:

https://www.gpxygpfx.com/CN/10.3964/j.issn.1000-0593(2023)10-3302-08 或 https://www.gpxygpfx.com/CN/Y2023/V43/I10/3302

[1] Liu Z, Lu Y, Peng Y, et al. Remote Sensing, 2019, 11(12): 1464.
[2] Wei L, Yuan Z, Yu M, et al. Sensors, 2019, 19(18): 3904.
[3] Tan K, Ma W, Chen L, et al. Journal of Hazardous Materials, 2021, 401: 123288.
[4] Jiang G, Zhou S, Cui S, et al. Sensors, 2020, 20(21): 6325.
[5] LIN Nan, LIU Han-lin, MENG Xiang-fa, et al(林楠, 刘翰霖, 孟祥发, 等). Transactions of the Chinese Society for Agricultural Machinery(农业机械学报), 2021, 52(3): 218.
[6] Wei L, Pu H, Wang Z, et al. Sensors, 2020, 20(14): 4056.
[7] MARHABA Turhun, MAMATTURSUM Eziz, WANG Wei-wei(麦尔哈巴·图尔贡, 麦麦提吐尔逊·艾则孜, 王维维). Journal of Ecology and Rural Environment(生态与农村环境学报), 2020, 36(12): 1626.
[8] CHENG Yong-sheng, ZHOU Yao(成永生, 周瑶). The Chinese Journal of Nonferrous Metals(中国有色金属学报), 2021, 31(11): 3450.
[9] LIU Yan-ping, LUO Qing, CHENG He-fa(刘彦平, 罗晴, 程和发). Journal of Agro-Environment Science(农业环境科学学报), 2020, 39(12): 2699.
[10] XIAO Ye-hui, SONG Ni-di, MENG Pan-pan, et al(肖烨辉, 宋妮迪, 孟盼盼, 等). Remote Sensing for Natural Resources(自然资源遥感), 2021, 33(4): 143.
[11] Wei L, Yuan Z, Zhong Y, et al. Applied Sciences, 2019, 9(9): 1943.
[12] FENG Shuai, CAO Ying-li, XU Tong-yu, et al(冯帅, 曹英丽, 许童羽, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2020, 40(8): 2584.
[13] Wu T, Yu J, Lu J, et al. Agriculture, 2020, 10(7): 292.
[14] JIA Ping-ping, SHANG Tian-hao, ZHANG Jun-hua, et al(贾萍萍, 尚天浩, 张俊华, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2020, 36(17): 125.
[15] 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.