基于知识驱动模型的遥感找矿预测方法研究

A study of the remote sensing ore prognostic method based on a data-driven model

摘要: 提要：本文根据ASTER遥感数据提供的蚀变矿物信息，结合典型矿床的找矿模型，在四川盐源地区进行了基于知识驱动模型的遥感找矿预测方法研究。利用地理信息系统的多种逻辑模型，对ASTER遥感数据解译出的蚀变矿物丰度及氧化物含量运用布尔逻辑，索引叠加和模糊逻辑等进行了GIS分析，最终确定出由ASTER遥感数据产生的找矿靶区，并在找矿靶区内得到了很好的野外验证效果。

Abstract: Abstract：Based on the alteration mineral information derived from the ASTER data, combined with a typical ore-prospecting model, the authors studied the ore prognostic method that utilized the data-driven model and remote sensing data obtained from Yanyuan area, Sichuan Province. Different logical models were used to analyze alteration mineral abundances and oxide contents on the basis of such means as Bool logic, overlay index and fuzzy logical method. As a result, the prospecting target was delineated, and the interpretation result was well verified in field work.