Geological Mapping via Convolutional Neural Network Based on Remote Sensing and Geochemical Survey Data in Vegetation Coverage Areas

Geological mapping in vegetation coverage areas is a challenging task. In this study, convolutional neural networks (CNNs) were employed for geological area based on remote sensing images and geochemical survey data. The Gram-Schmidt fusion technology was first applied to fuse Sentinel-2A ASTER enhance the spatial resolution enrich spectral information of fused then organically integrated with data according correlations between element contents reflectance objects. A case study six lithologic units Jilinbaolige, Inner Mongolia, China implemented using CNN model classification map obtained an overall accuracy 83.0%, which exhibited better performance contrast random forest (RF) model. results showed that CNNs can take full advantage features solve problems ‘salt pepper phenomenon’ against shallow machine learning algorithms, provide rich diagnostic mapping.