选矿智能摇床图像处理开发与应用  

Development and Application of Image Processing of the Intelligent Shaker for Mineral Processing

作　　者：张文康 刘丹[1,2,3] 杜钰 王春景 余龙舟 袁林逊 ZHANG Wen-kang;LIU Dan;DU Yu;WANG Chun-jing;YU Long-zhou;YUAN Lin-xun(Faculty of Land and Resources Engineering,Kunming University of Science and Technology;State Key Laboratory of Clean Utilization of Complex Nonferrous Metal Resources,Kunming University of Science and Technology;Yunnan Key Laboratory of Green Separationand Enrichment of Strategic Mineral Resources,Kunming University of Science and Technology;Yunnan Amde Electrical Engineering Co.,Ltd.;Yunnan Pinshi Intelligent Technology Co.,Ltd.)

机构地区：[1]昆明理工大学国土资源工程学院 [2]昆明理工大学省部共建复杂有色金属资源清洁利用国家重点实验室 [3]昆明理工大学云南省战略金属矿产资源绿色分离与富集重点实验室 [4]云南阿姆德电气工程有限公司 [5]云南品视智能科技有限公司

出　　处：《化工自动化及仪表》2023年第1期82-87,共6页Control and Instruments in Chemical Industry

摘　　要：为实现摇床的实时监测和自动调整接矿板的位置,采用机器视觉技术采集摇床实时矿带照片,研究了实时矿带图像识别算法,开发了摇床实时图像监测智能设备。工业试验表明,摇床实时图像监测智能设备能够自动采集矿带图片,图像识别算法实现了对摇床矿带带宽、矿带边界和矿带颜色特征的数字化解析,实现根据矿带变化实时自动调节接矿板的位置。该设备提升了选矿的各项指标,降低了人工成本。For purpose of realizing real-time monitoring of the shaker and automatically adjusting the position of ore connecting plates,the real-time ore belt photos of the shaker were collected by using machine vision technologies,and the real-time ore belt image recognition algorithm was studied,including the development of intelligent equipment for real-time image monitoring of the shaker.The industrial testing shows that,the shaker’s real-time image monitoring intelligent device can auto-collect ore bed images,and the image recognition algorithm can realize digital analysis of the shaking ore belt width,the ore belt and the ore belt color features,including auto-regulating position of the ore-connecting plates at real time according to the ore belt changes.This device can improve the indexes of mineral processing and reduce the labor cost.

关 键 词：图像处理 摇床 智能 机器学习 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]