摘要
土壤盐渍化是可溶性盐分在土壤中积累,导致土壤基本性质和质量恶化的过程,对农业生产和生态系统构成严重威胁。因此,精确监测土壤盐渍化的分布、面积、程度对指导区域盐渍化防治、保障农业生产安全具有重要意义。利用光学遥感卫星可以实现区域土壤盐分快速估测。然而基于高光谱遥感影像进行土壤盐分估测时,光谱受传感器、大气环境和土壤物理属性等多源因素影响,导致原始光谱存在噪声,从而使得估测模型精度下降。为此,本文采用正交信号校正(OSC)、分段直接标准化(PDS)和外部参数正交化(EPO)方法进行光谱校正处理,削弱光谱噪声,并利用随机森林模型构建土壤盐分估测模型。结果表明,经OSC、PDS和EPO算法校正处理后,提高了盐分含量与光谱的相关性,土壤盐分遥感估测精度提升明显,其中OSC表现效果最好,估测模型预测决定系数达到0.714,均方根误差为7.768 ms·cm-1。
Soil salinization is the accumulation of soluble salt in the soil,which leads to the deterioration of soil basic properties and quality,and poses a serious threat to agricultural production and ecosystem.Therefore,accurate monitoring of the distribution,area and degree of soil salinization is of great significance to guide the prevention and control of regional salinization and ensure the safety of agricultural production.Although optical remote sensing satellites enable rapid regional estimation of soil salinity,the accuracy of quantitative models based on hyperspectral imagery is often compromised.This is because the acquired spectral data are subject to multi-source interference from sensor noise,atmospheric effects,and variations in soil physical properties,which introduce noise into the original signal.Therefore,this study employed orthogonal signal correction(OSC),piecewise direct standardization(PDS),and external parameter orthogonalisation(EPO)for reducing spectral noise.A random forest model was then applied to develop a soil salinity estimation model.The results show that the calibration using OSC,PDS,and EPO algorithms enhanced the correlation between salt content and spectral data,thereby significantly improving the accuracy of soil salinity estimation via remote sensing.Among them,OSC demonstrated the best performance,achieving a coefficient of determination of 0.714 and a root mean square error of 7.768 ms·cm-1 for the estimation model.
作者
麻玮玮
林楠
吴梦红
张中帅
MAWeiwei;LIN Nan;WU Menghong;ZHANG Zhongshuai(College of Surveying and Exploration Engineering,Jilin Jianzhu University,Changchun Jilin 130118;Modern Industry College,Jilin Jianzhu University,Changchun Jilin 130118)
出处
《现代农业研究》
2025年第8期98-102,共5页
Modern Agriculture Research
基金
吉林省自然科学基金项目“基于UAV-TE技术的田块尺度下黑土土壤有机质快速估测研究”(项目编号:YDZJ202401507ZYTS)。
关键词
土壤盐分
高光谱遥感影像
光谱校正
正交信号校正
soil salinity
hyperspectral remote sensing imagery
spectral correction
orthogonal signal correction
