摘要
随着科学技术的不断发展,岩石学研究的方法和手段也不断丰富和更新。岩心矿物组分鉴定是岩石骨架、孔隙组合、填隙物类型等微观储层研究的主要依据,但由于岩心分析工作量大,传统分析方法面临效率低、精准度低、专家依赖度高、人才紧缺等挑战。文章在总结分析岩心矿物成分识别相关技术的基础上,利用语义分割网络模型,探索验证了一种基于岩心薄片图像的矿物成分特征识别技术,结合实例实现了矿物成分自动识别和组分占比统计。
With the continuous advancement of science and technology,the methods and means of petrological research are also constantly enriched and updated.The identification of mineral composition in rock cores is the main basis for micro-reservoir research,such as rock matrix,pore assemblage,and interstitial materials.However,due to the large workload of rock core analysis,traditional analysis methods face challenges such as low efficiency,low precision,high dependence on experts,and shortage of talents.In this context,on the basis of summarizing relevant technologies,this article explores and verifies a mineral composition feature recognition technology based on rock core thin section images using a semantic segmentation network model,and realizes the automatic recognition of mineral composition and component proportion ratio by combining with examples.
作者
姚皓骞
黄瑞
田榆杰
Yao Haoqian;Huang Rui;Tian Yujie(School of Earth Science and Resources,China University of Geosciences(Beijing),Beijing 100083;Tsinghua Sichuan Energy Internet Research Institute,Chengdu 610200)
出处
《中阿科技论坛(中英文)》
2024年第7期75-79,共5页
China-Arab States Science and Technology Forum
关键词
岩石薄片
矿物成分
语义分割
卷积神经网络
智能识别
Rock thin sections
Mineral composition
Semantic segmentation
Convolutional neural network
Intelligent identification