自动矿物学:Ⅱ.图像处理与自动测量  

Automated mineralogy:partⅡ.BSE images analysis and automated mineralogical characterization

作　　者：温利刚 贾木欣 王清 付强 赵建军 周俊武 WEN Ligang;JIA Muxin;WANG Qing;FU Qiang;ZHAO Jianjun;ZHOU Junwu(BGRIMM Technology Group,Beijing 102628,China;State Key Laboratory of Process Automation in Mining&Metallurgy,Beijing 102628,China;Beijing Key Laboratory of Process Automation in Mining&Metallurgy,Beijing 102628,China)

机构地区：[1]矿冶科技集团有限公司,北京102628 [2]矿冶过程自动控制技术国家重点实验室,北京102628 [3]矿冶过程自动控制技术北京市重点实验室,北京102628

出　　处：《中国矿业》2020年第S02期361-367,共7页China Mining Magazine

基　　金：国家科技支撑计划项目资助(编号:2012BAB19B01);国家自然科学基金项目资助(编号:51734005);矿冶科技集团有限公司青年科技创新基金项目资助(编号:04-1923)。

摘　　要：近年来,自动矿物学(或定量矿物学)技术改变了传统的以偏光显微镜为主的岩矿鉴定和矿石工艺参数测量的工作方式,已经成为工艺矿物学研究的重要手段。本文评述了基于扫描电镜的自动矿物学技术的图像处理、矿物自动测量的一些关键思路,主要包括:①选矿产品矿物定量分析的环氧树脂抛光片制备、自动矿物学技术基本原理;②矿石颗粒提取、矿物分相、X射线能谱采集等图像处理与自动测量过程;③矿物自动识别方法与矿物定量表征。在此基础上,总结了自动矿物学技术较体分析技术和传统人工手动逐粒分析方法的优势。以期为矿产资源可利用性评价和开发利用提供新的研究途径和思路。Automated mineralogy(AM),or described as quantitative mineralogy(QM),has been available with the scanning electron microscopy(SEM)-based automated image analysis system,which have been widely used in quantitative mineralogical investigation,especially the process mineralogical study.The authors discussed some key methods of SEM-based automated quantitative mineralogical technologies.These methods mainly included:the sample preparation of polished particle epoxy resin section of mineral processing products for automated mineralogical analysis,and the functional principle of the SEM-based automated mineralogy systems;BSE image analysis and automated measurement,such as the background removal,particulation,segmentation and automated EDS spectra acquisition,etc.;automated mineral identification methods and automated quantitative mineralogical characterization.On the basis,the advantages of automated mineralogical methods,compared with bulk analysis(e.g.,XRD,XRF,ICP-MS,ICP-AES,etc.)and traditional manual individual particle analysis using optical microscopy,were summarized.The authors hope that this paper can provide new approaches and ideas for evaluation and development of mineral resources.

关 键 词：矿物自动识别 BSE图像 EDS谱图 逐粒分析 矿物定量 自动矿物学 

分 类 号：TD912[矿业工程—选矿] P575[天文地球—矿物学]