基于极限学习机的遥感地球化学反演模型

地球化学勘查研究涉及大量采样工作，但在工作环境恶劣人们难以到达的地区，大范围、大比例尺的地球化学数据极难获取。本文基于极限学习机（ELM）构建遥感地球化学反演模型，弥补因为区域数据不足导致的找矿工作困难。依据偏最小二乘回归（PLSR）方法选取与地球化学数据相关性强的遥感影像成分，并根据极限学习机建立地球化学数据与遥感影像之间的非线性对应关系来获取未知地球化学异常，以此来指导找矿工作。实验中，选取研究区铜元素1：20万土壤地球化学数据与Landsat 8 OLI遥感影像进行反演分析。实验结果表明，基于ELM的遥感地球化学反演所取得的异常分布与已知矿点具有很好的对应度，验证了本文所提出模型的有效性。

Remote Sensing Geochemical Inversion Model by Using Extreme Learning Machine

Geochemical exploration research involves a large amount of sampling work, which is extremely difficult in inaccessible terrain with harsh working environments. The authors propose a geochemical inversion model with remote sensing images by using extreme learning machine (ELM) to alleviate the difficulty of ore prospecting in the areas with insufficient regional data. The partial least squares regression (PLSR) method is used to select the remote sensing image features which are highly correlated with geochemistry data. In this model, the nonlinear relationship between the geochemical data and the remote sensing images is established using ELM for getting unknown geochemical anomalies, after which the ore prospecting work can be further promoted. In the experiment, 1:200 000 soil geochemical data of Cu element and the Landsat 8 OLI remote sensing images were used for the inversion analysis. The experimental results showed that the anomalous distribution obtained by the ELM-based inversion model had a good correspondence with known ore spots, which verified the effectiveness of the proposed model.