煤岩显微组分自动识别技术研究进展  

Research progress on automatic coal maceral identification

作　　者：王越 白向飞 曲思建 WANG Yue;BAI Xiangfei;QU Sijian(School of Chemical and Environmental Engineering,China University of Mining and Technology(Beijing),Beijing 100083,China;Beijing Research Institute of Coal Chemistry,China Coal Research Institute Corporation Ltd.,Beijing 100013,China;State Key Laboratory of Coal Mining and Clean Utilization,Beijing 100013,China)

机构地区：[1]中国矿业大学(北京)化学与环境工程学院,北京100083 [2]煤炭科学技术研究院有限公司煤化工分院,北京100013 [3]煤炭资源高效开采与洁净利用国家重点实验室,北京100013

出　　处：《煤质技术》2021年第1期49-57,共9页Coal Quality Technology

基　　金：煤炭科学技术研究院有限公司科技发展基金资助项目(2018CX01)。

摘　　要：煤岩显微组分的形态特征多样,组分鉴定具有较强的专业性,需要测定者具有深厚的煤岩学基础并熟练运用偏光显微镜。传统组分鉴定存在主观性,测试时间也较长。煤岩显微组分自动识别形成了基于显微光度计和基于图像分析2种技术路线。基于显微光度计的识别方法硬件成本相对较低,通过点扫描或线扫描将显微组分的反射率值转化为电信号并通过数学处理方法实现自动识别。图像分析方法综合显微组分的反射率(灰度)和形貌参数进行自动识别,识别精度更高。随着图像处理技术和图像识别算法的发展,可实现基于显微组分的纹理特征的自动分类。显微组分的智能识别是煤岩显微组分自动识别技术的发展趋势,应在充分分析不同煤的显微组分特征的基础上开发针对性的特征提取和分类算法,以实现煤岩显微组分自动识别。It is highly professional of macerals identification as diversity morphologies have been found of coal macerals.The operators should have profound basis of coal petrology,and operators should be proficient to operate the polarizing microscope.The traditional maceral identification method is subjective and time-consuming.There are 2 automatic coal maceral identification techniques:the microphotometer-based technique and the image analysis method.The microphotometer-based technique has relatively low hardware cost,it transforms the maceral reflectivity into electrical signals through spot or line scanning,and automatic identification by mathematical treatment is achieved.The image analysis method combines the maceral reflectivity(gray value)and morphological parameters to automatic identify macerals,which enjoy higher recognition accuracy.With the development of image processing techniques and image identification algorithmics,the automatic maceral classification will be achieved based on the textural features.The intelligent identification of maceral is the future of automatic coal maceral identification.After fully analysis the characteristics of macerals in different coals,a specific feature extraction and classification method should be developed to achieve automatic coal maceral identification.

关 键 词：煤岩 显微组分 自动识别技术 图像分析 识别方法 智能识别 形貌参数 

分 类 号：TQ531[化学工程—煤化学工程]