基于YOLOv5的煤矿井下瓦斯钻杆智能识别方法  

Intelligent Recognition Method for Gas Drill Rods in Coal Mine Underground Based on YOLOv5

作　　者：蒋博文 张若楠 徐平安 谢玉麒 JIANG Bowen;ZHANG Ruonan;XU Pingan;XIE Yuqi(Pingan Coal Mining Engineering Technology Research Institute Co.,Ltd.,Huainan 232001,China;Huainan Mining(Group)Co.,Ltd.,Huainan 232000,China)

机构地区：[1]平安煤炭开采工程技术研究院有限责任公司,安徽淮南232001 [2]淮南矿业(集团)有限责任公司,安徽淮南232000

出　　处：《现代信息科技》2025年第6期142-145,共4页Modern Information Technology

摘　　要：随着煤炭开采深度的增加,瓦斯突出成为制约煤炭开采的重要因素。煤矿井下瓦斯钻杆计数对井下瓦斯抽采具有重要的指导意义。目前,主要使用的传统人工计数方式较为费时费力。为解决上述问题,提出一种基于YOLOv5的煤矿井下瓦斯钻杆智能识别方法。该方法通过目标检测算法YOLOv5对钻杆特征进行充分提取和聚合,并输出最终检测结果。通过多角度布置摄像头采集钻杆数据,构建了BOLT-1000钻杆数据集,并搭建了实验平台进行训练和验证。实验结果表明,该方法在自制数据集上具有较高的精确度和鲁棒性,能够准确识别井下瓦斯钻杆。With the increase of coal mining depth,gas outburst has become an important factor restricting coal mining.Counting gas drill rods in coal mines underground has important guiding significance for underground gas extraction.At present,the main used traditional manual counting method is more time-consuming and laborious.To solve the above problems,an intelligent identification method of underground gas drill rods based on YOLOv5 is proposed.This method fully extracts and aggregates the drill rod features through the object detection algorithm YOLOv5 and outputs the final detection results.The BOLT-1000 drill rod dataset is constructed by collecting drill rod data through multi-angle camera arrangement,and the experimental platform is built for training and verification.The experimental results show that the method has high accuracy and robustness on the self-made dataset,and can accurately identify the underground gas drill rods.

关 键 词：目标检测 瓦斯抽采 钻杆识别 YOLOv5 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]