Research on quantitative inversion of ion adsorption type rare earth ore based on convolutional neural network

Rare earth resource is a national strategic resource, which plays an essential role in the field of high technology research and development. In this paper, we aim to use remote sensing quantitative inversion prospecting technology, surface-to-surface mode, model evaluation through convolutional neural network achieve new method for large-scale, low-cost, rapid efficient exploration ion-adsorbed rare ore. The results show that RE 2 O 3 content samples has significant negative correlation with second, third fourth band GF-2 image, but no first image; convolution can be used reconstruct content. distribution map obtained by similar geochemical map, indicates invert sampling area. characteristics ion adsorption ore study area are basically consistent actual situation; there two main anomaly areas I known mining area, II prospective type deposit. It shows deposit based on Convolutional Neural Networks (CNN) feasible.