基于FastICA盲源分离法去除土壤干扰的小麦生物量高光谱估算

生态学杂志

基于FastICA盲源分离法去除土壤干扰的小麦生物量高光谱估算

李燕丽^1,2,吴士文^1,2,刘娅¹,王昌昆¹,刘杰^1,2,徐爱爱^1,2,潘贤章^1*#br#

（¹土壤与农业可持续发展国家重点实验室，中国科学院南京土壤研究所，南京 210008； ^2中国科学院大学, 北京 100049）

出版日期:2017-04-10 发布日期:2017-04-10

Applying fast independent component analysis algorithm of blind source separation method to remove soil effects on hyperspectral data for wheat biomass estimation.

LI Yan-li^1,2, WU Shi-wen^1,2, LIU Ya¹, WANG Chang-kun¹, LIU Jie^1,2, XU Ai-ai^1,2, PAN Xian-zhang^1*#br#

(¹ State Key Laboratory of Soil and Sustainable Agriculture, Institute of Soil Science, Chinese Academy of Sciences, Nanjing 210008, China; ^2 University of Chinese Academy of Sciences, Beijing 100049, China).

Online:2017-04-10 Published:2017-04-10

摘要/Abstract

摘要： 高光谱技术是一种快速无损监测植被生物量的有效方法，但土壤背景的干扰一直是生物量监测的主要限制因素之一。本研究试图利用盲源分离（blind source separation, BSS）法分离出净植被光谱，达到消除土壤背景影响，提高小麦生物量估算精度的目的。本研究对110组小麦冠层光谱数据进行快速独立分量分析（fast independent component analysis，FastICA）处理，提取净植被光谱，并对比了FastICA处理前后所建的偏最小二乘回归（partial least squares regression, PLSR）模型估算精度。结果表明：FastICA算法可有效分离土壤光谱和植被光谱；且基于净植被光谱建立的小麦生物量估算模型精度得到明显提升，建模集RPD_c（ratio of performance to deviation of the calibration）和交叉验证集RPD_cv（ratio of performance to deviation of the cross calibration）分别由原始光谱的1.83和1.64提高至2.77和2.09；可见，FastICA可以作为有效的光谱数据预处理方法，显著提高小麦生物量的估算精度，为利用遥感技术进行大尺度、精准监测生物量提供了方法支持和理论依据。

关键词: 土壤碳矿化, 降雨减少, 土壤团聚体

Abstract: Hyperspectral technique has been an effective method to monitor the vegetation biomass as a rapid and nondestructive approach. However, the accuracy of biomass estimation is always limited by the influence of soil background. The purpose of this study aimed to alleviate the effects of soil on spectra and improve the accuracy of wheat biomass estimation based on the extracted vegetation spectra by blind source separation (BSS) method. In this study, with the application of fast independent component analysis (FastICA), pure vegetation spectra were extracted from the 110 groups of original fieldobserved canopy spectra, and the wheat biomass estimation accuracy were compared before and after FastICA with the partial least squares regression (PLSR). The results showed that the FastICA method could separate the soil spectra and vegetation spectra effectively, and the accuracy of wheat biomass estimation was significantly improved based on the extracted vegetation spectra, as compared with the original spectral, with the improvement of the ratio of performance to deviation of the calibration (RPD_c) and the ratio of performance to deviation of the cross calibration (RPD_cv) from 1.83 and 1.64 to 2.77 and 2.09, respectively. These results indicated that FastICA method could be applied as an effective spectral preprocessing method to significantly improve the accuracy of biomass estimation, thus providing guidance for accurate regional monitoring of wheat biomass by hyperspectral technology.

Key words: soil carbon mineralization, precipitation reduction, soil aggregate.

李燕丽1,2,吴士文1,2,刘娅1,王昌昆1,刘杰1,2,徐爱爱1,2,潘贤章1*. 基于FastICA盲源分离法去除土壤干扰的小麦生物量高光谱估算[J]. 生态学杂志.

LI Yan-li1,2, WU Shi-wen1,2, LIU Ya1, WANG Chang-kun1, LIU Jie1,2, XU Ai-ai1,2, PAN Xian-zhang1*. Applying fast independent component analysis algorithm of blind source separation method to remove soil effects on hyperspectral data for wheat biomass estimation.[J]. cje.

[1]	何婉莹, 谷思玉, 谷邵臣, 车延静, 王紫颖, 袁望博, 杨佳宇. 深翻-旋耕轮耕与有机肥配施对农田黑土团聚体内酶活性的影响 [J]. 生态学杂志, 2023, 42(5): 1090-1098.
[2]	王斌, 王汝振, 李甜, 张玉革, 姜勇. 大气沉降氮在土壤-植物系统中的留存机制 [J]. 生态学杂志, 2023, 42(2): 463-470.
[3]	唐志伟, 周文涛, 王勃然, 钟康裕, 熊瑞, 龙攀, 徐莹, 傅志强. 垄厢栽培对水稻根系特性与土壤物理性质的影响 [J]. 生态学杂志, 2021, 40(12): 3961-3969.
[4]	张宇, 蒋代华, 黄金兰, 王明释, 邓华为. 粉垄耕作对赤红壤团聚体粒级分布和稳定性的影响 [J]. 生态学杂志, 2021, 40(12): 3922-3932.
[5]	丁俊男, 于少鹏, 史传奇, 王慧. 寒区湿地不同土地利用方式对土壤理化性质和团聚体稳定性的影响 [J]. 生态学杂志, 2021, 40(11): 3543-3551.
[6]	李仁杰, 杨立琼, 庄杰, 石亚楠, 陈希娟. 有机质和团聚体对双酚A迁移行为的影响 [J]. 生态学杂志, 2020, 39(12): 4148-4156.
[7]	朱源山,王义东,郭长城,尚云涛,薛冬梅,李军,王中良. 天津盐碱化沼泽湿地开垦对土壤团聚体有机与无机碳含量的影响 [J]. 生态学杂志, 2020, 39(1): 206-216.
[8]	宋莉群,王义东,李冬初,张会民,娄翼来,窦渤凯,尚云涛,王中良. 长期退耕对红壤团聚体碳氮磷生态化学计量特征的影响 [J]. 生态学杂志, 2019, 38(6): 1707-1715.
[9]	叶思源,陈展,曹吉鑫,尚鹤. 模拟酸雨和接种外生菌根真菌对马尾松土壤养分、土壤团聚体及有机碳组分的影响 [J]. 生态学杂志, 2019, 38(4): 1141-1148.
[10]	张倩,韩贵琳,柳满,阳昆桦,刘强. 贵州普定喀斯特关键带土壤磷分布特征及其控制因素 [J]. 生态学杂志, 2019, 38(2): 321-328.
[11]	谢尧,赵琼,李炎真,于占源,林贵刚,曾德慧. 干旱化对樟子松固沙林氮磷循环的影响 [J]. 生态学杂志, 2019, 38(12): 3593-3600.
[12]	江淼华,倪梦颖,周嘉聪,陈岳民,杨玉盛. 增温和降雨减少对杉木幼林土壤酶活性的影响 [J]. 生态学杂志, 2018, 37(11): 3210-3219.
[13]	郭鑫炜1,2,李志远1,2,秦胜金1,2,周永斌1,2,魏亚伟1,2*. 林农复合对沙地樟子松人工林土壤团聚体、有机碳及微生物生物量碳分配的影响 [J]. 生态学杂志, 2017, 36(11): 3036-3042.
[14]	杨宁1,2,邹冬生2*,付美云1,杨满元1,林仲桂1. 紫色土丘陵坡地恢复中土壤团聚体特征及其与土壤性质的关系 [J]. 生态学杂志, 2016, 35(9): 2361-2368.
[15]	周玉燕1,2,贾晓红2,3**,张烜铭1,马进琴1,陈凤鸣1,高艳红2. 土壤碳矿化潜力对沙坡头人工固沙植被演变的响应 [J]. 生态学杂志, 2013, 32(6): 1371-1377.