基于野外实测光谱的玄武岩地球化学成分反演

发布时间：2018-03-17 09:51

本文选题：玄武岩　切入点：地球化学成分　出处：《国土资源遥感》2017年01期 　论文类型：期刊论文

【摘要】：岩石地球化学成分对岩石分类、成因及演化研究有重要地质意义。利用遥感手段反演岩石地球化学成分是一个较新的课题,也是遥感相关应用研究的难点。以甘肃省柳园镇南部二叠纪玄武岩带为研究目标,在进行系统采样、光谱分析、地球化学测试的基础上,采用偏最小二乘回归(partial least squares regression,PLSR)对拥有2 150个波段的实测波谱数据及相应的6种主量矿物数据进行反演建模。首先选择有效的预处理方法对目标数据集进行优化,再利用k折交叉检验方法获得最小均方根误差下的最适主成分个数。运算结果表明,PLSR模型具有较好的稳定性和精度,在利用遥感数据进行岩石地球化学成分反演方面有很好的应用前景。
[Abstract]:Rock geochemical composition of rock classification has important geological significance of Genesis and evolution. By means of remote sensing inversion of rock geochemical composition is a new topic, but also related to the application of remote sensing research difficulty. In Gansu Province, south of the town of Willow Park as the research object with the Permian basalts, in systematic sampling, spectrum analysis based on the geochemical test, using partial least squares regression (partial least squares regression, PLSR) inverse modeling of measured spectral data with 2150 bands and 6 corresponding major mineral data. First select the effective pretreatment method to optimize the target data set, and then use the optimal number of principal components K fold cross validation method to obtain the minimum root mean square error. The computational results show that the PLSR model has better precision and stability, rock geochemistry in using remote sensing data There is a good application prospect in the inversion of the study components.

【作者单位】：中国国土资源航空物探遥感中心;中国地质大学(北京)地球科学与资源学院地质过程与矿产资源国家重点实验室;
【基金】：国土资源部公益性行业科研基金项目“月球试验场遥感探测研究”(编号:201211071-2) 中国地质调查局地质调查项目“高光谱地质调查技术方法研究”(编号:12120115040801)共同资助
【分类号】：P627

【参考文献】