山东煤炭科技2025，Vol.43Issue(9)：198-202,207,6.DOI:10.3969/j.issn.1005-2801.2025.09.041
基于图像识别的胶带载煤量计算方法
Calculation Method of Belt Transportation Volume Based on Image Recognition
郭力源 1王延辉2
作者信息
- 1. 国家矿山安全监察局山西局,山西 太原 030012
- 2. 精英数智科技股份有限公司,山西 太原 030006
- 折叠
摘要
Abstract
Aiming at the problem that the measurement method of belt scales in coal mine production monitoring is prone to human interference,leading to uncontrollable accuracy,this paper proposes a non-contact calculation method for belt coal transportation volume based on image recognition.The YOLOv8 instance segmentation model is used to accurately extract the coal flow contour.Combined with perspective transformation correction and dynamic angle of repose calculation model,the real-time measurement of the cross-sectional area of coal flow is realized.Field tests in Tashan Coal Mine of Jinneng Holding Group show that the daily accuracy of the belt production recognition reaches 92.95%.This method provides a technically innovative and practically reliable technical solution for intelligent monitoring of coal mine production.关键词
胶带载煤量监测/实例分割/透视校正/堆积角模型Key words
belt transportation volume monitoring/instance segmentation/perspective correction/angle of repose model分类
矿业与冶金引用本文复制引用
郭力源,王延辉..基于图像识别的胶带载煤量计算方法[J].山东煤炭科技,2025,43(9):198-202,207,6.
