ISSN 1008-5548

CN 37-1316/TU

2021年27卷  第4期
<返回第4期

基于目标检测网络的煤矸石识别

Identification of coal and gangue based on object detection network

高新宇, 李 博, 王璐瑶, 李廉洁, 王学文

(太原理工大学机械与运载工程学院; 煤矿综采装备山西省重点试验室, 山西太原 030024)


DOI:10.13732/j.issn.1008-5548.2021.04.010

收稿日期: 2021-04-23, 修回日期:2021-05-01,在线出版时间:2021-06-01 09:55。

基金项目:山西省重点研发计划项目,编号:201903D121074。

第一作者简介:高新宇(1995—),男,硕士研究生,研究方向为煤矸石智能分选。E-mail: 813358884@qq.com。

通信作者简介:王学文(1979—),男,博士,教授,博士生导师,研究方向为机械设计及理论。E-mail: wxuew@163.com。


摘要:按照实际工况在实验室搭建煤矸分选平台,采用深层目标检测网络对煤矸石进行在线识别,根据分选时煤矸石的形状和大小,将目标检测网络中的特征金字塔设定为3个尺度,并确定锚(参考边界框)的形状和大小;比较10个IOU(Intersection-over-union)阈值下验证集的平均精度(AP),并在煤矸石分选平台对目标检测网络进行动态测试。结果表明:IOU为0.8时,目标检测网络的分类和定位效果最佳,动态识别的精确度和召回率均达到95%以上。

关键词:煤矸分选;图像处理;深度学习;目标检测

Abstract: A platform of coal-gangue separation was set up in the laboratory according to the actual working conditions. Adopting object detection network to identify coal and gangue online and according to the shape and size of coal gangue during sorting, the feature pyramid in the object detection network was set as three scales, and the shape and size of the anchor were determined. Comparing the AP (average precision) of the validation set under 10 IOU (intersection over union) thresholds, and a dynamic test was carried out on the separation platform built. The results show that the classification and positioning effect of the object detection network was the best when the IOU was 0.8, and the precision and recall of dynamic identification can reach more than 95%.

Keywords: coal-gangue separation; image processing; deep learning; object detection


参考文献(References):

[1]国家能源局. 国家能源局关于印发《煤炭清洁高效利用行动计划(2015—2020年)》的通知:国能煤炭[2015]141号[A/OL]. (2015-04-27)[2021-04-20].http://zfxxgk.nea.gov.cn/auto85/201505/t20150505_1917.htm.

[2]HOU W. Identification of coal and gangue by feed-forward neural network based on data analysis[J]. International Journal of Coal Preparation and Utilization, 2019, 39(1): 33-43.

[3]ZHAO Y, HE X. Recognition of coal and gangue based on X-ray[J]. Applied Mechanics and Materials, 2013, 275-277: 2350-2353.

[4]程学珍, 王伟, 卫阿盈. 煤与矸石在线分选系统设计[J]. 工矿自动化, 2013, 39(8): 95-98.

[5]MCCOY J T, AURET L. Machine learning applications in minerals processing: a review[J]. Minerals Engineering, 2019, 132: 95-109.

[6]徐浩然, 贺福强, 薛亚军, 等. 基于遗传算法的多功能木质板材配比[J]. 中国粉体技术, 2021, 27(2): 30-36.

[7]李曼, 段雍, 曹现刚, 等. 煤矸分选机器人图像识别方法和系统[J]. 煤炭学报, 2020, 45(10): 3636-3644.

[8]李良晖. 放顶煤工作面煤矸混合度自动识别研究进展[J]. 煤炭工程, 2017, 49(10): 30-34.

[9]田妍, 田丰. 放顶煤开采过程煤矸识别技术发展现状及前景[J]. 煤炭工程, 2018, 50(10): 142-145.

[10]张慧, 王坤峰, 王飞跃. 深度学习在目标视觉检测中的应用进展与展望[J]. 自动化学报, 2017, 43(8): 1289-1305.

[11]雷亚国, 贾峰, 周昕, 等. 基于深度学习理论的机械装备大数据健康监测方法[J]. 机械工程学报, 2015, 51(21): 49-56.

[12]徐志强, 吕子奇, 王卫东, 等. 煤矸智能分选的机器视觉识别方法与优化[J]. 煤炭学报, 2020, 45(6): 2207-2216.

[13]GAO R, SUN Z, LI W, et al. Automatic coal and gangue segmentation using U-net based fully convolutional networks[J]. Energies, 2020, 13(4): 1-13.

[14]LI D, ZHANG Z, XU Z, et al. An image-based hierarchical deep learning framework for coal and gangue detection[J]. IEEE Access, 2019, 7: 184686-184699.

[15]黎玲利, 孟令兵, 李金宝. 多尺度特征提取和多级别特征融合的显著性目标检测方法[J]. 工程科学与技术, 2021, 53(1): 170-177.

[16]SZEGEDY C, LIU W, JIA Y, et al. Going deeper with convolutions[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston: IEEE, 2015: 1-9.