基于SD−YOLOv5s−4L的煤矿井下无人驾驶电机车多目标检测

赵伟; 王爽; 赵东洋

doi:10.13272/j.issn.1671-251x.2023070100

基于SD−YOLOv5s−4L的煤矿井下无人驾驶电机车多目标检测

doi: 10.13272/j.issn.1671-251x.2023070100

赵伟^{1, 2,},
王爽^{1, 2, 3, ,},
赵东洋^{1, 2}

1.
安徽理工大学深部煤矿采动响应与灾害防控国家重点实验室，安徽淮南　232001
2.
安徽理工大学机械工程学院，安徽淮南　232001
3.
矿山智能技术与装备省部共建协同创新中心，安徽淮南　232001

基金项目: 国家自然科学基金项目（52274152）；安徽省高校杰出青年科研项目（2022AH020056）。

详细信息

作者简介:
赵伟（2000—），男，安徽桐城人，硕士研究生，研究方向为矿山智能化装备与技术，E-mail：zhaow201105@126.com

通讯作者:
王爽（1991—），女，安徽马鞍山人，副教授，博士，研究方向为煤矿机器人，E-mail：chk0519@126.com。

中图分类号: TD64
计量
- 文章访问数: 1261
- HTML全文浏览量: 28
- PDF下载量: 56
- 被引次数: 0
出版历程
- 收稿日期: 2023-07-28
- 修回日期: 2023-11-17
- 网络出版日期: 2023-11-27

Multi object detection of underground unmanned electric locomotives in coal mines based on SD-YOLOv5s-4L

ZHAO Wei^{1, 2
,},
WANG Shuang^{1, 2, 3
, ,},
ZHAO Dongyang^{1, 2}

1.
State Key Laboratory of Mining Response and Disaster Prevention and Control in Deep Coal Mines, Anhui University of Science and Technology, Huainan 232001, China
2.
School of Mechanical Engineering, Anhui University of Science and Technology, Huainan 232001, China
3.
Collaborative Innovation Center for Mining Intelligent Technology and Equipment, Huainan 232001, China

摘要

摘要: 为解决煤矿井下无人驾驶电机车由于光照不均、高噪声等复杂环境因素导致的多目标检测精度低及小目标识别困难问题，提出一种基于SD−YOLOv5s−4L的煤矿井下无人驾驶电机车多目标检测模型。在YOLOv5s基础上进行以下改进，构建SD−YOLOv5s−4L网络模型：引入SIoU损失函数来解决真实框与预测框方向不匹配的问题，使得模型可以更好地学习目标的位置信息；在YOLOv5s头部引入解耦头，增强网络模型的特征融合与定位准确性，使得模型可以快速捕捉目标的多尺度特征；引入小目标检测层，将原三尺度检测层增至4层，以增强模型对小目标的特征提取能力和检测精度。在矿井电机车多目标检测数据集上进行实验，结果表明：SD−YOLOv5s−4L网络模型对各类目标的平均精度均值（mAP）为97.9%，对小目标的平均检测精度（AP）为98.9%，较YOLOv5s网络模型分别提升了5.2%与9.8%；与YOLOv7，YOLOv8等其他网络模型相比，SD−YOLOv5s−4L网络模型综合检测性能最佳，可为实现矿井电机车无人驾驶提供技术支撑。
- 井下无人驾驶 /
- 电机车 /
- 多目标检测 /
- YOLOv5s /
- SIoU /
- 解耦头 /
- 小目标检测
Abstract: Due to complex environmental factors such as uneven illumination and high noise, unmanned electric locomotives in coal mines have low accuracy in multi object detection and difficulty in recognizing small objects. In order to solve the above problems, a multi object detection model for underground unmanned electric locomotives in coal mines based on SD-YOLOv5s-4L is proposed. On the basis of YOLOv5s, the following improvements are made to construct the SD-YOLOv5s-4L network model. The model introduces the SIoU loss function to solve the problem of mismatch between the direction of the real box and the predicted box, so that the model can better learn the position information of the object. The model introduces decoupled heads at the head of YOLOv5s to enhance the feature fusion and positioning accuracy of the network model. It enables the model to quickly capture multi-scale features of the object. The model introduces a small object detection layer to increase the original three scale detection layer to four scale. It enhances the model's feature extraction capability and detection precision for small objects. The experiment is conducted on a multi object detection dataset of the mine electric locomotives. The results show the following points. The mean average precision (mAP) of the SD-YOLOv5s-4L network model for various types of objects is 97.9%, and the average precision (AP) for small objects is 98.9%. Compared with the YOLOv5s network model, it improves by 5.2% and 9.8%, respectively. Compared with other network models such as YOLOv7 and YOLOv8, the SD-YOLOv5s-4L network model has the best comprehensive detection performance and can provide technical support for achieving unmanned driving of the mine electric locomotives.
- underground unmanned driving /
- electric locomotive /
- multi object detection /
- YOLOv5s /
- SIoU /
- decoupling head /
- small object detection

HTML全文

图 1 YOLOv5s网络结构

Figure 1. YOLOv5s network structure

下载: 全尺寸图片幻灯片

图 2 解耦头结构

Figure 2. Decoupled head structure

下载: 全尺寸图片幻灯片

图 3 SD−YOLOv5s−4L网络结构

Figure 3. SD-YOLOv5s-4L network structure

下载: 全尺寸图片幻灯片

图 4 部分数据集图像

Figure 4. Partial dataset images

下载: 全尺寸图片幻灯片

图 5 不同网络模型检测效果对比

Figure 5. Comparison of detection results of different algorithms

下载: 全尺寸图片幻灯片

图 6 各组消融实验mAP对比

Figure 6. Comparison of mAP in each ablation experiment

下载: 全尺寸图片幻灯片

图 7 不同网络模型的mAP对比

Figure 7. Comparison of mAP for different network models

下载: 全尺寸图片幻灯片

表 1 实验环境

Table 1. Experimental environment

实验环境	参数
操作系统	Ubuntu 18.04
CPU	Intel（R） Xeon（R） Platinum 8350C CPU@2.6 GHz
GPU	RTX 3090（24 GiB）
深度学习框架	PyTorch 1.9.0
编程语言	Python 3.8
CUDA	11.1

下载: 导出CSV

表 2 超参数设置

Table 2. Hyper-parameter setting

参数名称	数值
batch-size	32
momentum	0.937
decay	0.0005
learning rate	0.01
epochs	301

下载: 导出CSV

表 3 消融实验结果

Table 3. Ablation experiment results

实验	网络模型		AP		mAP
实验	网络模型	person	signal light	stone	mAP
1	YOLOv5s	0.945	0.945	0.891	0.927
2	YOLOv5s+ SIoU	0.946	0.942	0.931	0.940
3	YOLOv5s+解耦头	0.956	0.950	0.923	0.943
4	YOLOv5s+小目标检测层	0.972	0.959	0.971	0.968
5	YOLOv5s+SIoU+ 小目标检测层+解耦头	0.980	0.967	0.989	0.979

下载: 导出CSV

表 4 对比实验结果

Table 4. Comparative experimental results

实验	网络模型		AP		F₁	mAP
实验	网络模型	person	signal light	stone	F₁	mAP
1	YOLOv5n	0.900	0.913	0.923	0.89	0.912
2	YOLOv5m	0.965	0.960	0.935	0.93	0.954
3	YOLOv5s	0.945	0.945	0.891	0.91	0.927
4	YOLOv7	0.967	0.950	0.921	0.94	0.946
5	YOLOv8	0.948	0.926	0.985	0.92	0.953
6	SD−YOLOv5s−4L	0.980	0.967	0.989	0.96	0.979

下载: 导出CSV

参考文献(23)

[1]	王国法,赵国瑞,任怀伟. 智慧煤矿与智能化开采关键核心技术分析[J]. 煤炭学报,2019,44(1):34-41. doi: 10.13225/j.cnki.jccs.2018.5034 WANG Guofa,ZHAO Guorui,REN Huaiwei. Analysis on key technologies of intelligent coal mine and intelligent mining[J]. Journal of China Coal Society,2019,44(1):34-41. doi: 10.13225/j.cnki.jccs.2018.5034
[2]	孙继平. 煤矿智能化与矿用5G[J]. 工矿自动化,2020,46(8):1-7. SUN Jiping. Coal mine intelligence and mine-used 5G[J]. Industry and Mine Automation,2020,46(8):1-7.
[3]	于骞翔,张元生. 井下电机车轨道障碍物图像处理方法的智能识别技术[J]. 金属矿山,2021(8):150-157. doi: 10.19614/j.cnki.jsks.202108024 YU Qianxiang,ZHANG Yuansheng. Track obstacle intelligent recognition technology of mine electric locomotive based on image processing[J]. Metal Mine,2021(8):150-157. doi: 10.19614/j.cnki.jsks.202108024
[4]	韩江洪,卫星,陆阳,等. 煤矿井下机车无人驾驶系统关键技术[J]. 煤炭学报,2020,45(6):2104-2115. HAN Jianghong,WEI Xing,LU Yang,et al. Driverless technology of underground locomotive in coal mine[J]. Journal of China Coal Society,2020,45(6):2104-2115.
[5]	胡青松,孟春蕾,李世银,等. 矿井无人驾驶环境感知技术研究现状及展望[J]. 工矿自动化,2023,49(6):128-140. doi: 10.13272/j.issn.1671-251x.18115 HU Qingsong,MENG Chunlei,LI Shiyin,et al. Research status and prospects of perception technology for unmanned mining vehicle driving environment[J]. Journal of Mine Automation,2023,49(6):128-140. doi: 10.13272/j.issn.1671-251x.18115
[6]	李程,车文刚,高盛祥. 一种用于航拍图像的目标检测算法[J]. 山东大学学报(理学版),2023,58(9):59-70. LI Cheng,CHE Wengang,GAO Shengxiang. A object detection algorithm for aerial images[J]. Journal of Shandong University(Natural Science),2023,58(9):59-70.
[7]	GIRSHICK R,DONAHUE J,DARRELL T,et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]. 2014 IEEE Conference on Computer Vision and Pattern Recognition,Columbus,2014:580-587.
[8]	HE Kaiming,GKIOXARI G,DOLLAR P,et al. Mask R-CNN[C]. 2017 IEEE International Conference on Computer Vision,Venice,2017:2980-2988.
[9]	REDMON J,DIVVALA S,GIRSHICK R,et al. You only look once:unified,real-time object detection[C]. IEEE Conference on Computer Vision and Pattern Recognition,Las Vegas,2016:779-788.
[10]	LIU Wei,ANGUELOV D,ERHAN D,et al. SSD:single shot multi-box Detector[C]. European Conference on Computer Vision,2016:21-37.
[11]	李伟山,卫晨,王琳. 改进的Faster RCNN煤矿井下行人检测算法[J]. 计算机工程与应用,2019,55(4):200-207. doi: 10.3778/j.issn.1002-8331.1711-0282 LI Weishan,WEI Chen,WANG Lin. Improved faster RCNN approach for pedestrian detection in underground coal mine[J]. Computer Engineering and Applications,2019,55(4):200-207. doi: 10.3778/j.issn.1002-8331.1711-0282
[12]	HE Deqiang,LI Kai,CHEN Yanjun,et al. Obstacle detection in dangerous railway track areas by a convolutional neural network[J]. Measurement Science and Technology,2021,32(10). DOI: 10.1088/1361-6501/abfdde.
[13]	HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al. Identity mappings in deep residual networks[C]. European Conference on Computer Vision,2016:630-645.
[14]	郝帅,张旭,马旭,等. 基于CBAM−YOLOv5的煤矿输送带异物检测[J]. 煤炭学报,2022,47(11):4147-4156. HAO Shuai,ZHANG Xu,MA Xu,et al. Foreign object detection in coal mine conveyor belt based on CBAM-YOLOv5[J]. Journal of China Coal Society,2022,47(11):4147-4156.
[15]	郑玉珩,黄德启. 改进MobileViT与YOLOv4的轻量化车辆检测网络[J]. 电子测量技术,2023,46(2):175-183. ZHENG Yuheng,HUANG Deqi. Lightweight vehicle detection network based on MobileViT and YOLOv4[J]. Electronic Measurement Technology,2023,46(2):175-183.
[16]	杨艺,付泽峰,高有进,等. 基于深度神经网络的综采工作面视频目标检测[J]. 工矿自动化,2022,48(8):33-42. YANG Yi,FU Zefeng,GAO Youjin,et al. Video object detection of the fully mechanized working face based on deep neural network[J]. Journal of Mine Automation,2022,48(8):33-42.
[17]	葛淑伟,张永茜,秦嘉欣,等. 基于优化SSD−MobileNetV2的煤矿井下锚孔检测方法[J]. 采矿与岩层控制工程学报,2023,5(2):66-74. GE Shuwei,ZHANG Yongqian,QIN Jiaxin,et al. Rock bolt borehole detection method for underground coal mines based on optimized SSD-MobileNetV2[J]. Journal of Mining and Strata Control Engineering,2023,5(2):66-74.
[18]	GEVORGYAN Z. SIoU loss:more powerful learning for bounding box regression[EB/OL]. [2023-05-12]. https://arxiv.org/abs/2205.12740.
[19]	ZHENG Zhaohui,WANG Ping,REN Dongwei,et al. Enhancing geometric factors in model learning and inference for object detection and instance segmentation[J]. IEEE Transactions on Cybernetics,2021,52(8):8574-8586.
[20]	WU Yue,CHEN Yinpeng,YUAN Lu,et al. Rethinking classification and localization for object detection[C]. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition,Seattle,2020:10183-10192.
[21]	LIN T-Y,DOLLAR P,GIRSHICK R B,et al. Feature pyramid networks for object detection[C]. 2017 IEEE Conference on Computer Vision and Pattern Recognition,Honolulu,2017:936-944.
[22]	LIU Shu,QI Lu,QIN Haifang,et al. Path aggregation network for instance segmentation[C]. IEEE/CVF Conference on Computer Vision and Pattern Recognition,Salt Lake City,2018:8759-8768.
[23]	KARAKAYA M,CELEBI M F,GOK A E,et al. Discovery of agricultural diseases by deep learning and object detection[J]. Environmental Engineering and Management Journal,2022,21(1):163-173. doi: 10.30638/eemj.2022.016