黄磊,陶明,刘玉龙,徐源泉,向恭梁

• 1:中广核铀业发展有限公司
• 2:中南大学资源与安全工程学院

摘要(Abstract)：

为解决矿山爆破块度识别效率低下、精度不足以及复杂环境干扰等难题，提出了一种基于双目视觉的爆破块度识别新方法。通过构建YOLOv8实例分割模型，精准提取复杂光照条件下的爆后矿岩轮廓。结合双目测量技术，基于三维坐标转换与视差计算原理，获取碎石的最大尺寸。搭建了室内实验平台，验证了块度识别方法在不同参数下的碎石识别与尺寸计算精度。最后，提出了露天矿爆破块度智能化识别技术架构，开发了块度自动识别与分析系统。室内模拟测试结果表明：较低的拍摄高度有助于提高模型的识别精度，碎石间接触虽对个别目标识别造成影响，但整体精度未显著下降，所有碎石的识别准确率均在85%以上。动态环境下的识别精度略有下降，但80%的碎石尺寸计算准确率仍保持在90%以上，整体误差仍控制在可接受范围内，满足爆破碎石实时监测及后续分析需求。该方法已成功应用于纳米比亚湖山矿，并结合射频识别(RFID)技术获取物料来源信息，实现爆区碎石块度分布(FSD)的动态监测、精确分析与全面评估，为露天台阶爆破效果评价提供了先进的技术手段。

关键词(KeyWords)： 双目视觉;智能识别系统;RFID;爆破块度

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金面上项目(52274105);; 湖南省研究生科研创新项目(CX20240247)~~

作者(Author): 黄磊,陶明,刘玉龙,徐源泉,向恭梁

参考文献(References)：

• [1] TAO M,XU Y Q,ZHAO R,et al.Energy control and block performance optimization of bench blasting[J].International Journal of Rock Mechanics and Mining Sciences,2024,180:105830.
• [2] 卢文波，孟婷，胡英国.岩石爆破破碎模拟和块度预报的研究现状与展望[J].工程爆破，2024,30(5):20-28.[2] LU Wen-bo,MENG Ting,HU Ying-guo.A review of numerical simulation of rock fragmentation by blasting and prediction of fragments[J].Engineering Blasting,2024,30(5):20-28.(in Chinese)
• [3] 李萍丰，王秀龙，闫小兵，等.柱状药包侧向爆破下抵抗线对岩石破碎特性的影响研究[J].岩土力学，2024,45(5):1388-1396.[3] LI Ping-feng,WANG Xiu-long,YAN Xiao-bing,et al.Effects of burden on rock crushing characteristics under lateral detonation of cylindrical charges[J].Rock and Soil Mechanics,2024,45(5):1388-1396.(in Chinese)
• [4] 杨晓妤.基于图像的岩体裂隙识别及爆破块度预测研究[D].天津：天津大学，2020.[4] YANG Shu-yu.Research on rock mass fracture identification and blasting fragmentation prediction based on image[D].Tianjin:Tianjin University,2020.(in Chinese)
• [5] 赵胜.地下矿山爆破块度智能识别与参数动态优化[D].沈阳：东北大学，2021.[5] ZHAO Sheng.Intelligent identification of blasting fragmentation and parameter dynamic optimization of blasting in underground mine[D].Shenyang:Northeastern University,2021.(in Chinese)
• [6] 王大坤，叶图强，闫大洋，等.基于三维激光扫描技术的爆破块度识别[J].有色金属工程，2022,12(11):120-127.[6] WANG Da-kun,YE Tu-qiang,YAN Da-yang,et al.Blasting fragmentation identification based on 3D laser scanning technology[J].Nonferrous Metals Engineering,2022,12(11):120-127.(in Chinese)
• [7] 刘俊伟，陈晓青.基于计算机视觉的爆破块度分布统计[J].矿业研究与开发，2024,44(1):190-196.[7] LIU Jun-wei,CHEN Xiao-qing.Statistics of blasting fragmentation distribution based on computer vision[J].Mining Research and Development,2024,44(1):190-196.(in Chinese)
• [8] 刘强，施富强，汪旭光，等.基于三维激光点云的爆破块度统计预测方法[J].煤炭学报，2020,45(S2):781-790.[8] LIU Qiang,SHI Fu-qiang,WANG Xu-guang,et al.Statistical estimation of blast fragmentation based on point cloud of muckpile[J].Journal of China Coal Society,2020,45(S2):781-790.(in Chinese)
• [9] 陈然，杨仕教，朱忠华，等.基于双门限阈值的爆破块度图像识别研究[J].工程爆破，2020,26(2):57- 64.[9] CHEN Ran,YANG Shi-jiao,ZHU Zhong-hua,et al.Research on image recognition of blasting block based on double threshold[J].Engineering Blasting,2020,26(2):57- 64.(in Chinese)
• [10] 雷振，张智宇，黄永辉，等.岩石爆破破碎能耗随抵抗线的变化规律[J].爆炸与冲击，2021,41(7):151-160.[10] LEI Zhen,ZHANG Zhi-yu,HUANG Yong-hui,et al.An investigation of energy consumption variation in rock blasting breaking with the resistance line[J].Explosion and Shock Waves,2021,41(7):151-160.(in Chinese)
• [11] 马泗洲，刘科伟，杨家彩，等.不耦合装药下岩石爆破块体尺寸的分布特征[J].爆炸与冲击，2024,44(4):122-140.[11] MA Si-zhou,LIU Ke-wei,YANG Jia-cai,et al.Size distribution characteristics of blast-induced rock fragmentation under decoupled charge structures[J].Explosion and Shock Waves,2024,44(4):122-140.(in Chinese)
• [12] 段继超，宗琦，汪海波，等.起爆顺序对露天台阶爆破效果影响的数值模拟及现场试验[J].中国安全生产科学技术，2023,19(5):109-116.[12] DUAN Ji-chao,ZONG Qi,WANG Hai-bo,et al.Numerical simulation and field test on influence of blasting sequence on open-pit bench blasting effect[J].Journal of Safety Science and Technology,2023,19(5):109-116.(in Chinese)
• [13] 黄雪峰，储亚坤，杨赛群，等.露天矿山台阶爆破合理微差时间优选及应用[J].煤矿爆破，2024,42(1):15-20.[13] HUANG Xue-feng,CHU Ya-kun,YANG Sai-qun,et al.Optimal selection and application of reasonable millisecond delay time for bench blasting in open-pit mines[J].Coal Mine Blasting,2024,42(1):15-20.(in Chinese)
• [14] 苗小虎，喻智，谭期仁，等.纳米比亚湖山铀矿空气间隔装药试验研究[J].中国矿业，2020,29(8):165-168.[14] MIAO Xiao-hu,YU Zhi,TAN Qi-ren,et al.Experimental study on air-deck charge of Husab uranium mine in Namibia[J].China Mining Magazine,2020,29(8):165-168.(in Chinese)
• [15] 罗志华，史秀志，陈飞，等.径向不耦合装药对粉矿控制的影响研究[J].爆破，2020,37(3):47-55.[15] LUO Zhi-hua,SHI Xiu-zhi,CHEN Fei,et al.Study on influence of radial decoupling charge on fine ore control[J].Blasting,2020,37(3):47-55.(in Chinese)
• [16] 叶海旺，彭瑞，田治平，等.基于孔间延时优化的骨料矿山台阶爆破大块率控制[J/OL].金属矿山，2024,1-10.[2025- 03-28].http://kns.cnki.net/kcms/detail/34.1055.td.20240812.1438.004.html.[16] YE Hai-wang,PENG Rui,TIAN Zhi-ping,et al.Large lump rate control of aggregate mine bench blasting based on inter-hole delay optimization[J/OL].Metal Mine,2024,1-10.[2025- 03-28].http://kns.cnki.net/kcms/detail/34.1055.td.20240812.1438.004.html.(in Chinese)
• [17] 杨学军，钟金彪，林开颜，等.双目立体视觉技术及其在智慧农业中的应用研究进展[J].农业工程学报，2025,41(1):27-39.[17] YANG Xue-jun,ZHONG Jin-biao,LIN Kai-yan,et al.Research progress on binocular stereo vision technology and its applications in smart agriculture[J].Transactions of the Chinese Society of Agricultural Engineering,2025,2025,41(1):27-39.(in Chinese)
• [18] 张婵.基于自适应双阈值的智能双目视觉系统设计[J].网络安全技术与应用，2023(11):57-59.[18] ZHANG Chan.Design of an intelligent binocular vision system based on adaptive dual thresholds[J].Network Security Technology & Application,2023(11):57-59.(in Chinese)
• [19] 陈丘轲.基于改进YOLOv8的钢材缺陷目标检测方法[D].上海：华东师范大学，2024.[19] CHEN Qiu-ke.A steel defect target detection methodbased on improved YOLOv8[D].Shanghai:East China Normal University,2024.(in Chinese)
• [20] 杜雨珊.基于YOLOv8的水下目标快速检测方法研究[D].大连：大连海事大学，2024.[20] DU Yu-shan.Research on rapid underwater target detection method based on YOLOv8[D].Dalian:Dalian Maritime University,2024.(in Chinese)
• [21] 刘振，董绍江，罗家元，等.基于双目视觉和改进YOLOv8n的火灾检测及测距方法[J].陕西科技大学学报，2025,43(1):152-160.[21] LIU Zhen,DONG Shao-jiang,LUO Jia-yuan,et al.Fire detection and ranging method based on binocular vision and improved YOLOv8n[J].Journal of Shaanxi University of Science & Technology,2025,43(1):152-160.(in Chinese)
• [22] 韩强.面向小目标检测的改进YOLOv8算法研究[D].长春：吉林大学，2023.[22] HAN Qiang.Research on improved YOLOV8 algorithm for small target detection[D].Changchun:Jilin University,2023.(in Chinese)