Zircon classification from cathodoluminescence images using deep learning  

作　　者：Dongyu Zheng Sixuan Wu Chao Ma Lu Xiang Li Hou Anqing Chen Mingcai Hou 

机构地区：[1]State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation,Chengdu University of Technology,Chengdu 610051,PR China [2]Key Laboratory of Deep-time Geography&Environment Reconstruction and Applications of Ministry of Natural Resources&Institute of Sedimentary Geology,Chengdu University of Technology,Chengdu 610051,PR China [3]Program of Geology and Geophysics,Missouri University of Science and Technology,Rolla 65401,USA [4]College of Earth Sciences,Chengdu University of Technology,Chengdu 610051,PR China

出　　处：《Geoscience Frontiers》2022年第6期111-121,共11页地学前缘（英文版）

基　　金：supported by the IUGS Deep-time Digitial Earth (DDE) Big Science Program;National Natural Science Foundation of China (Grant Nos. 42050104, 42172137, 41888101, and 42102092);Open Fund (Grant No. PLC20211102) of State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation of Chengdu University of Technology;the Everest Scientific Research Program of Chengdu University of Technology (Grant No. 2021ZF11402)。

摘　　要：Zircon is a widely-used heavy mineral in geochronological and geochemical research because it can extract important information to understand the history and genesis of rocks. Zircon has various types,and an accurate examination of zircon type is a prerequisite procedure before further analysis.Cathodoluminescence(CL) imaging is one of the most reliable ways to classify zircons. However, current CL image examination is conducted by manual work, which is time-consuming, bias-prone, and requires expertise. An automated and bias-free method for zircon classification is absent but necessary. To this end, deep convolutional neural networks(DCNNs) and transfer learning are applied in this study to classify the common types of zircons, i.e., igneous, metamorphic, and hydrothermal zircons. An atlas with over 4000 CL images of these three types of zircons is created, and three DCNNs are trained using these images. The results of this study indicate that the DCNNs can distinguish hydrothermal zircons from other zircons, as indicated by the highest accuracy of 100%. Although similar textures in igneous and metamorphic zircons pose great challenges for zircon classification, the DCNNs successfully classify 95% igneous and 92% metamorphic zircons. This study demonstrates the high accuracy of DCNNs in zircon classification and presents the great potentiality of deep learning techniques in numerous geoscientific disciplines.

关 键 词：ZIRCON Cathodoluminescence image Deep learning Transfer learning 

分 类 号：P597.3[天文地球—地球化学]