摘要
为进一步优化煤矿井下皮带机皮带破损问题的检测效果,在已有研究基础上,提出一种整合机器视觉和线激光辅助的检测方案,并以YOLOv3智能网络模型为基础,整合Dense-Darknet53特征提取网络、Adam优化器等智能技术,设计智能检测方法。结果表明,该检测方法在准确率上更具优势,有效规避了将划痕检测为破损的失误问题,具有一定应用价值。
To further optimize the detection effect of belt damage in coal mine underground belt conveyors,a detection scheme integrating machine vision and line-laser assistance is proposed based on existing research.An intelligent detection method is designed by incorporating intelligent technologies such as the YOLOv3 intelligent network model,the Dense-Darknet53 feature extraction network,and the Adam optimizer.The results indicate that this detection method demonstrates superior accuracy and effectively avoids the misclassification of scratches as damage,thus exhibiting certain application value.
作者
宋海明
SONG Haiming(Shanxi Kaijia Energy Group Co.,Ltd.,Jinzhong,Shanxi 032000,China)
出处
《自动化应用》
2025年第15期115-117,121,共4页
Automation Application
关键词
皮带破损
智能检测
检测方法
belt damage
intelligent detection
detection method
