铜业工程Issue(2)：38-47,10.DOI:10.3969/j.issn.1009-3842.2025.02.005
面向智能光电分选的矿石粒度在线识别
Online Recognition of Ore Particle Size for Intelligent Photoelectric Sorting Technology
摘要
Abstract
In the field of non-ferrous metal mining,high-throughput multi-grade intelligent photoelectric sorting is crucial for improving resource utilization and achieving sustainable development.However,X-ray transmission-based sorting systems face challenges due to the coupling effect of ore thickness and average atomic number,which reduces decision-making accuracy.By utilizing non-contact sensing methods to online recognize the particle size of moving ores,it is expected to decouple ore thickness information from X-ray transmission images,which serves as a critical breakthrough for intelligent sorting technology in large-scale multi-grade ore processing.Aiming at the application demands of online identification of ore particle size,this paper reviews potential feasible technical methods based on two-dimensional images,three-dimensional point clouds,and multimodal data fusion technology.Futhermore,the advantages and disadvantages of existing online ore particle size identification technologies and their respective limitations are analyzed.Finally,the technical development direction and technical route for online identification of ore particle size in ore photoelectric sorting systems are proposed.This study provides theoretical insights and technical pathways for innovation in photoelectric ore sorting technologies.关键词
智能光电分选/粒度在线识别/矿石图像分割/矿石点云/多模态融合Key words
intelligent photoelectric sorting/online ore particles recognition/ore image segmentation/ore point cloud/multimodal fu-sion分类
矿山工程引用本文复制引用
陈串串,刘祖鹏,严昊,徐阳,肖罡..面向智能光电分选的矿石粒度在线识别[J].铜业工程,2025,(2):38-47,10.基金项目
国家重点研发计划项目(2023YFC3904200),江西省自然科学基金项目(20224ACB218002),江西省重大科技研发专项项目(20232ACE01010)资助 （2023YFC3904200）
