基于机器视觉的带式输送机高精度煤流检测研究

季现亮; 张文杰; 王玉强; 刘勇; 田祖织; 付拯

doi:10.13272/j.issn.1671-251x.2024030028

基于机器视觉的带式输送机高精度煤流检测研究

doi: 10.13272/j.issn.1671-251x.2024030028

1.
临沂矿业集团菏泽煤电有限公司郭屯煤矿，山东菏泽　274000
2.
中国矿业大学机电工程学院，江苏徐州　221116
3.
山东能源集团鲁西矿业有限公司，山东菏泽　274000

基金项目: 国家自然科学基金资助项目（52375069）。

详细信息

作者简介:
季现亮（1972—），男，山东齐河人，工程师，研究方向为智能矿山技术，E-mail：997476358@qq.com

通讯作者:
张文杰（1998—），男，江苏徐州人，硕士研究生，研究方向为智能矿山技术，E-mail: 714849265@qq.com。

中图分类号: TD634
计量
- 文章访问数: 111
- HTML全文浏览量: 19
- PDF下载量: 26
- 被引次数: 0
出版历程
- 收稿日期: 2024-03-12
- 修回日期: 2024-05-26
- 网络出版日期: 2024-06-13

Research on high-precision coal flow detection of belt conveyors based on machine vision

1.
Guotun Coal Mine, Linyi Mining Group Heze Coal and Electricity Co., Ltd., Heze 274000, China
2.
School of Mechanical and Electrical Engineering, China University of Mining and Technology, Xuzhou 221116, China
3.
Shandong Energy Group Luxi Mining Co., Ltd., Heze 274000, China

摘要

摘要: 针对现有基于机器视觉的带式输送机煤流检测方法存在的图像细节缺失、在多处断裂或断裂间距较大区域拟合效果较差的问题，基于直射斜收式激光三角测量原理，提出了一种基于机器视觉的带式输送机高精度煤流检测系统，将线激光发射器布置在带式输送机测量位置正上方并垂直照射煤堆，煤堆随带式输送机匀速运动，利用相机在斜上方实时拍摄包含激光条纹的煤堆表面图像。对煤流检测系统进行标定，包括相机内参数标定和激光平面标定，得到煤堆的高度信息；对煤流截面激光条纹图像进行处理，从提取精度、算法实时性等角度对比分析了灰度重心法和区域骨架法，根据对比结果选用区域骨架法提取激光条纹中心；针对利用图像膨胀操作进行激光条纹断裂修补拟合效果较差的问题，提出采用最小二乘法作为激光条纹断裂修补算法，相较于闭运算，最小二乘法拟合处理的平滑效果更好，精度较高；建立煤流截面积计算模型，通过计算每一帧上煤堆的横截面积，即可得出不同带速下的煤流体积。实验结果表明，当带速分别为0.25，0.5，1 m/s时，煤流检测系统误差均较小，最大误差分别为2.78%，3.61%和3.89%，验证了煤流检测系统具有较高的准确性。
- 带式输送机 /
- 煤流检测 /
- 机器视觉 /
- 线激光发射 /
- 激光三角测量 /
- 激光条纹中心提取 /
- 激光条纹断裂修补 /
- 区域骨架法 /
- 最小二乘法
Abstract: In response to the problems of missing image details and poor fitting effect in multiple fractures or areas with large fracture spacing in existing machine vision based coal flow detection methods for belt conveyors, a high-precision coal flow detection system for belt conveyors based on machine vision is proposed. It is based on the principle of direct beam oblique collection laser triangulation. The line laser emitter is arranged directly above the measurement position of the belt conveyor and vertically irradiates the coal pile. The coal pile moves uniformly with the belt conveyor, and a camera at an oblique angle is used to capture real-time images of the surface of the coal pile containing laser stripes. The method calibrates the coal flow detection system, including camera internal parameter calibration and laser plane calibration, to obtain the height information of the coal pile. The processing of laser stripe images on coal flow cross-sections is carried out. The gray center of gravity method and regional skeleton method are compared and analyzed from multiple perspectives such as extraction precision and algorithm real-time performance. Based on the comparison results, the regional skeleton method is selected to extract the center of laser stripes. Aiming at the problem of poor fitting effect of laser stripe fracture repair using image dilation operation, the least squares method is proposed as the laser stripe fracture repair algorithm. Compared with closed operations, the least squares method has better smoothing effect and higher precision in fitting processing. The method establishes a coal flow cross-sectional area calculation model. By calculating the cross-sectional area of the coal pile at each frame, the coal flow volume at different belt speeds can be obtained. The experimental results show that when the belt speeds are 0.25, 0.5, and 1 m/s respectively, the detection system errors are relatively small, with maximum errors of 2.78%, 3.61%, and 3.89%. It verifies that the coal flow detection system has high accuracy.
- belt conveyor /
- coal flow detection /
- machine vision /
- linear laser emission /
- direct beam oblique collection laser triangulation /
- laser stripe center extraction /
- laser stripe fracture repair /
- regional skeleton method /
- least squares method

HTML全文

图 1 直射斜收式激光三角测量原理

Figure 1. Principle of direct oblique receiving laser triangulation

下载: 全尺寸图片幻灯片

图 2 煤流检测系统硬件平台

Figure 2. Coal flow detection system hardware platform

下载: 全尺寸图片幻灯片

图 3 标定相机内参时采集的图像

Figure 3. Images captured during camera calibration

下载: 全尺寸图片幻灯片

图 4 畸变校正前后对比

Figure 4. Comparison before and after distortion correction

下载: 全尺寸图片幻灯片

图 5 低位置图像

Figure 5. Low-position images

下载: 全尺寸图片幻灯片

图 6 高位置图像

Figure 6. High-position images

下载: 全尺寸图片幻灯片

图 7 红色通道分离前后激光条纹

Figure 7. Laser stripes before and after red channel separation

下载: 全尺寸图片幻灯片

图 8 图像滤波前后激光条纹

Figure 8. Laser stripes before and after image filtering

下载: 全尺寸图片幻灯片

图 9 阈值分割激光条纹

Figure 9. Threshold segmentation of laser stripe

下载: 全尺寸图片幻灯片

图 10 激光条纹灰度分布

Figure 10. Laser stripe gray distribution

下载: 全尺寸图片幻灯片

图 11 激光条纹中心线提取结果

Figure 11. Laser stripe centerline extraction results

下载: 全尺寸图片幻灯片

图 12 形态学断裂修补激光曲线

Figure 12. Morphological fracture repair laser curve

下载: 全尺寸图片幻灯片

图 13 最小二乘法激光曲线拟合

Figure 13. Least square method laser curve fitting

下载: 全尺寸图片幻灯片

图 14 闭运算和最小二乘法拟合效果对比

Figure 14. Comparison between closed operations and least square method

下载: 全尺寸图片幻灯片

图 15 煤流截面积计算模型

Figure 15. Calculation model of coal flow cross-section

下载: 全尺寸图片幻灯片

图 16 激光条纹像素点序号与x方向真实世界坐标对应关系

Figure 16. Correspondence between the pixel number of laser stripes and the real-world coordinates in the x direction

下载: 全尺寸图片幻灯片

图 17 煤流体积检测模型

Figure 17. Coal flow volume detection model

下载: 全尺寸图片幻灯片

图 18 标准模型

Figure 18. Standard model

下载: 全尺寸图片幻灯片

图 19 标准模型激光条纹拟合曲线

Figure 19. Standard model laser stripe fitting curves

下载: 全尺寸图片幻灯片

图 20 标准模型世界坐标

Figure 20. Standard model world coordinates

下载: 全尺寸图片幻灯片

表 1 带式输送机相关参数

Table 1. Belt conveyor related parameters

参数	值
额定运量/（t·h⁻¹）	15
运输距离/m	2.5
胶带宽度/mm	400
胶带速度/（m·s⁻¹）	0～1

下载: 导出CSV

表 2 相机内参信息

Table 2. Camera internal reference information

参数	值
平均误差	0.046 038
单个像元宽度/μm	4.836 41
单个像元高度/μm	4.8
焦距/mm	15.486 9
畸变参数/m⁻²	−658.29
中心点横坐标/像素	598.431
中心点纵坐标/像素	491.725
图像宽度/像素	1 280
图像高度/像素	1 024

下载: 导出CSV

表 3 中心线提取算法对比

Table 3. Comparison of centerline extraction algorithms

算法	提取精度级别	用时/ms	方向性	断裂拟合	中心线形状改变
灰度重心法	亚像素	21.8	差	断裂小处拟合	小
区域骨架法	亚像素	10.8	好	不拟合	小

下载: 导出CSV

表 4 闭运算和最小二乘法性能对比

Table 4. Comparison of performance between closed operations and least square method

拟合方法	拟合效果	提取精度	提取速度/ms	复杂度
闭运算	差	亚像素	2.171	简单
最小二乘法	好	亚像素	3.082	较复杂

下载: 导出CSV

表 5 标准模型检测结果

Table 5. Standard model detection results

尺寸/cm	宽度/cm	宽度误差/%	高度/cm	高度误差/%	面积/cm²	面积误差/%
5×5	5.029	0.58	4.9585	−0.83	24.937	−0.25
8×8	8.055	0.69	8.0046	0.06	64.476	0.74
10×8	10.050	0.50	7.9909	−0.11	80.318	0.40
10×10	10.060	0.60	10.080 0	0.80	101.360	1.36

下载: 导出CSV

表 6 煤堆体积测量结果

Table 6. Measurement results of coal pile volume

实际体积/cm³	带速为0.25 m/s		带速为0.5 m/s		带速为1 m/s
实际体积/cm³	测量值/cm³	误差/%	测量值/cm³	误差/%	测量值/cm³	误差/%
295.4	301.6	2.09	302.0	2.21	304.9	3.22
498.0	503.7	1.14	508.9	2.19	514.8	3.37
617.4	633.0	2.53	639.7	3.61	640.5	3.74
725.7	745.9	2.78	749.4	3.27	753.9	3.89
773.4	792.5	2.47	794.6	2.73	795.4	2.84
956.5	957.0	0.05	968.5	1.26	973.5	1.78

下载: 导出CSV

参考文献(24)

[1]	李晓丽,王庆福. 基于GA−BP神经网络的带式输送机故障监测系统研究[J]. 煤炭技术,2021,40(12):222-224. LI Xiaoli,WANG Qingfu. Research on fault monitoring system of belt conveyor based on GA-BP neural network[J]. Coal Technology,2021,40(12):222-224.
[2]	宋钦一. 基于故障树和贝叶斯网络的带式输送机故障诊断[J]. 矿山机械,2022,50(9):55-58. doi: 10.3969/j.issn.1001-3954.2022.09.013 SONG Qinyi. Fault diagnosis on belt conveyor based on fault tree and Bayesian network[J]. Mining & Processing Equipment,2022,50(9):55-58. doi: 10.3969/j.issn.1001-3954.2022.09.013
[3]	GAN Hong,CHEN Kun,ZHONG Xinghong. Static analysis on the measurement system of an electronic belt scale[J]. Applied Mechanics and Materials,2013,345:525-529. doi: 10.4028/www.scientific.net/AMM.345.525
[4]	赵学军,杨征,闫雪. 基于语义分割的带式输送机煤料运输区域检测算法[J]. 计算机应用与软件,2024,41(2):56-61,99. doi: 10.3969/j.issn.1000-386x.2024.02.008 ZHAO Xuejun,YANG Zheng,YAN Xue. Coal transportation area detection algorithm of belt conveyor based on semantic segmentation[J]. Computer Applications and Software,2024,41(2):56-61,99. doi: 10.3969/j.issn.1000-386x.2024.02.008
[5]	张文科,郭瑜,赵辉. 基于图像识别的煤矿带式输送机自适应调速系统设计[J]. 煤炭工程,2024,56(1):220-224. ZHANG Wenke,GUO Yu,ZHAO Hui. Self-adaptive speed control system of coal mine conveying belt based on image recognition[J]. Coal Engineering,2024,56(1):220-224.
[6]	汪心悦,乔铁柱,庞宇松,等. 基于TOF深度图像修复的输送带煤流检测方法[J]. 工矿自动化,2022,48(1):40-44,63. WANG Xinyue,QIAO Tiezhu,PANG Yusong,et al. Coal flow detection method for conveyor belt based on TOF depth image restoration[J]. Industry and Mine Automation,2022,48(1):40-44,63.
[7]	周富林,黄靖. AI边缘计算在工业视觉识别系统中的应用[J]. 现代传输,2022(6):54-56. doi: 10.3969/j.issn.1673-5137.2022.06.009 ZHOU Fulin,HUANG Jing. Application of AI edge computing in industrial visual recognition system[J]. Modern Transmission,2022(6):54-56. doi: 10.3969/j.issn.1673-5137.2022.06.009
[8]	李纪栋,蒲绍宁,翟超,等. 基于视频识别的带式输送机煤量检测与自动调速系统[J]. 煤炭科学技术,2017,45(8):212-216. LI Jidong,PU Shaoning,ZHAI Chao,et al. Coal quantity detection and automatic speed regulation system of belt conveyor based on video identification[J]. Coal Science and Technology,2017,45(8):212-216.
[9]	杨春雨,顾振,张鑫,等. 基于深度学习的带式输送机煤流量双目视觉测量[J]. 仪器仪表学报,2021,41(8):164-174. YANG Chunyu,GU Zhen,ZHANG Xin,et al. Binocular vision measurement of coal flow of belt conveyors based on deep learning[J]. Chinese Journal of Scientific Instrument,2021,41(8):164-174.
[10]	曾飞,吴青,初秀民,等. 带式输送机物料瞬时流量激光测量方法[J]. 湖南大学学报(自然科学版),2015,42(2):40-47. ZENG Fei,WU Qing,CHU Xiumin,et al. Measurement of material instantaneous flow on belt conveyors based on laser scanning[J]. Journal of Hunan University(Natural Sciences),2015,42(2):40-47.
[11]	胡而已. 融合激光扫描与机器视觉的煤流量测量研究[J]. 煤炭工程,2021,53(11):146-151. HU Eryi. Coal flow measurement based on laser scanning and machine vision[J]. Coal Engineering,2021,53(11):146-151.
[12]	李淑军,田昌勇,周传扬. 基于结构光的激光自动焊接跟踪研究[J]. 应用光学,2023,44(6):1280-1285. doi: 10.5768/JAO202344.0610016 LI Shujun,TIAN Changyong,ZHOU Chuanyang. Research on automatic laser welding tracking based on structured light[J]. Journal of Applied Optics,2023,44(6):1280-1285. doi: 10.5768/JAO202344.0610016
[13]	朱铮涛,裴炜冬,李渊,等. 基于远心镜头的激光三角测距系统研究与实现[J]. 激光与光电子学进展,2018,55(3):191-196. ZHU Zhengtao,PEI Weidong,LI Yuan,et al. Research and implementation of laser triangulation system based on telecentric lens[J]. Laser & Optoelectronics Progress,2018,55(3):191-196.
[14]	祝磊,韩自营,阮宇静,等. 基于机器视觉的轮胎胎面检测系统设计与实现[J]. 计算机工程与设计,2018,39(6):1782-1787. ZHU Lei,HAN Ziying,RUAN Yujing,et al. Design and implementation of tire tread detection system based on machine vision[J]. Computer Engineering and Design,2018,39(6):1782-1787.
[15]	郑彬,罗山,蒋银成. 基于RGB图像处理的轮胎胎面缺陷检测方法研究[J]. 制造业自动化,2023,45(6):35-38,49. doi: 10.3969/j.issn.1009-0134.2023.06.007 ZHENG Bin,LUO Shan,JINAG Yincheng. Research on tire tread defect detection method based on RGB image processing[J]. Manufacturing Automation,2023,45(6):35-38,49. doi: 10.3969/j.issn.1009-0134.2023.06.007
[16]	唐义杰,胡超,张倚玮,等. 强椒盐噪声下的模糊边缘自适应中值滤波算法[J]. 电子制作,2022,30(16):89-91. doi: 10.3969/j.issn.1006-5059.2022.16.027 TANG Yijie,HU Chao,ZHANG Yiwei,et al. Fuzzy edge adaptive median filtering algorithm under strong salt and pepper noise[J]. Practical Electronics,2022,30(16):89-91. doi: 10.3969/j.issn.1006-5059.2022.16.027
[17]	刘传洋,吴一全. 基于红外图像的电力设备识别及发热故障诊断方法研究进展 [J/OL]. 中国电机工程学报:1-27[2024-04-12]. http://kns.cnki.net/kcms/detail/11.2107.TM.20240226.0957.002.html. LIU Chuanyang,WU Yiquan. Research progress on power equipment identification and heating fault diagnosis methods based on infrared images[J/OL]. Proceedings of the CSEE:1-27[2024-04-12]. http://kns.cnki.net/kcms/detail/11.2107.TM.20240226.0957.002.html.
[18]	宋立彬,张淑艳. 基于机器视觉的煤流量快速检测方法[J]. 煤炭技术,2023,42(9):241-243. SONG Libin,ZHANG Shuyan. Rapid detection method of coal flow based on machine vision[J]. Coal Technology,2023,42(9):241-243.
[19]	吴玉波,陈迪来,杨超,等. 基于Steger算法的多线结构光中心提取[J]. 应用激光,2023,43(10):188-195. WU Yubo,CHEN Dilai,YANG Chao,et al. Multi-line structured light center extraction based on improved steger algorithm[J]. Applied Laser,2023,43(10):188-195.
[20]	陈哲,王生怀,钟明. 3D线激光相机的激光条纹中心提取方法[J]. 工具技术,2023,57(10):155-160. doi: 10.3969/j.issn.1000-7008.2023.10.032 CHEN Zhe,WANG Shenghuai,ZHONG Ming. Laser stripe center extraction method for 3D line laser cameras[J]. Tool Engineering,2023,57(10):155-160. doi: 10.3969/j.issn.1000-7008.2023.10.032
[21]	胡石. 基于激光扫描的工业机器人焊接焊缝跟踪方法[J]. 重庆科技学院学报(自然科学版),2023,25(5):69-75. doi: 10.3969/j.issn.1673-1980.2023.05.013 HU Shi. Study on welding seam tracking method of industrial robot based on laser scanning[J]. Journal of Chongqing University of Science and Technology (Natural Sciences Edition),2023,25(5):69-75. doi: 10.3969/j.issn.1673-1980.2023.05.013
[22]	王浩,赵小辉,徐龙哲,等. 结构光视觉辅助焊接的轨迹识别与控制技术[J]. 焊接学报,2023,44(6):50-57,132. doi: 10.12073/j.hjxb.20220715002 WANG Hao,ZHAO Xiaohui,XU Longzhe,et al. Research on trajectory recognition and control technology of structured light vision-assisted welding[J]. Transactions of the China Welding Institution,2023,44(6):50-57,132. doi: 10.12073/j.hjxb.20220715002
[23]	南诺,侯作勋,曹东晶,等. 一种基于图像形态学的深空图像模糊复原方法[J]. 航天返回与遥感,2023,44(2):101-108. doi: 10.3969/j.issn.1009-8518.2023.02.011 NAN Nuo,HOU Zuoxun,CAO Dongjing,et al. Deep-air image blur restoration method based on image morphology[J]. Spacecraft Recovery & Remote Sensing,2023,44(2):101-108. doi: 10.3969/j.issn.1009-8518.2023.02.011
[24]	田江云,温欣,刘旭东,等. 基于改进最小二乘法的椭圆形玻璃幕墙缺陷多传感器融合检测技术[J]. 无损检测,2024,45(1):28-32. TIAN Jiangyun,WEN Xin,LIU Xudong,et al. Multi-sensor fusion detection technology for elliptical glass curtain wall defects based on improved least squares method[J]. Nondestructive Testing,2024,45(1):28-32.