摘要
在油气勘探、评价及开发中,岩性识别和薄片鉴定是十分重要的基础工作,准确的薄片识别结果可以为勘探和开发提供可靠的依据。传统的人工判定方法或实验室分析方法具有主观性强、效率低、自动化程度低等问题。目前基于内容的智能图像识别技术在准确性和具体应用方面还面临着许多难题。论文基于国内外相关研究成果与油气勘探与开发中岩芯薄片图像的特点及要求,设计并研制成功薄片图像自动识别系统和薄片智能鉴定系统。利用图像梯度分布和色彩分析进行火成岩岩石薄片智能分类,对所有像素进行类别划分进而得到整体的鉴定结果,实现了省时、高效、高精度的薄片智能鉴定成果。
In oil and gas exploration,evaluation and development,lithology identification and thin sections identification are very important basic tasks,and accurate thin sections identification results can provide a reliable basis for exploration and development.Traditional manual judgment methods or laboratory analysis methods have problems such as strong subjectivity,low efficiency,and low degree of automation.At present,content-based intelligent image recognition technology still faces many difficulties in terms of accuracy and specific applications.Based on the relevant research results at home and abroad and the characteristics and requirements of core sections images in oil and gas exploration and development,the paper successfully designed and developed a thin section image automatic recognition system and a thin section intelligent identification system.Using image gradient distribution and color analysis to intelligently classify igneous rock thin sections,all pixels are classified to obtain the overall identification result,which realizes time-saving,efficient,and high-precision intelligent identification of thin sections.
作者
曹蒙
王志章
李冰涛
曲康
裴升杰
贾小玉
CAO Meng;WANG Zhizhang;LI Bingtao;QU Kang;PEI Shengjie;JIA Xiaoyu(China University of Petroleum(Beijing)School of Earth Sciences,Beijing,102249;State Key Laboratory of Petroleum Resources and Exploration,China University of Petroleum(Beijing),Beijing,102249)
出处
《地质论评》
CAS
CSCD
北大核心
2023年第4期1581-1588,共8页
Geological Review
关键词
火成岩
深度学习
图像梯度
色彩分析
薄片智能鉴定
igneous rock
deep learning
image gradient
color analysis
intelligent identification of thin sections