Foreign object detection of coal mine underground conveyor belt based on Stair-YOLOv7-tiny
-
摘要: 针对现有煤矿井下输送带异物检测方法应对复杂场景适应性差、无法满足实时性和轻量化要求、处理尺寸差异较大异物时表现不佳的问题,基于轻量化YOLOv7−tiny模型进行改进,提出了一种Stair−YOLOv7−tiny模型,并将其用于煤矿井下输送带异物检测。该模型在高效层聚合网络(ELAN)模块中添加特征拼接单元,形成阶梯ELAN(Stair−ELAN)模块,将不同层级的低维特征与高维特征进行融合,加强了特征层级间的直接联系,提升了信息捕获能力,增强了模型对不同尺度目标和复杂场景的适应性;针对检测头引入阶梯特征融合(Stair−fusion),形成阶梯检测头(Stair−head)模块,通过逐层融合不同分辨率的检测头特征,增强了中低分辨率检测头的特征表达能力,实现了特征信息的互补。实验结果表明:Stair−YOLOv7−tiny模型在输送带异物开源数据集CUMT−BelT上的检测效果优于CBAM−YOLOv5,YOLOv7−tiny及其轻量化模型,准确率、平均精度均值、召回率和精确率分别达98.5%,81.0%,82.2%和88.4%,检测速度为192.3帧/s;在某矿井下输送带监控视频分析中,Stair−YOLOv7−tiny模型未出现漏检或误检,实现了输送带异物的准确检测。
-
关键词:
- 输送带异物检测 /
- YOLOv7−tiny /
- 多尺度目标检测 /
- Stair−fusion /
- 高效层聚合网络 /
- 检测头
Abstract: The existing methods for detecting foreign objects in underground coal mine conveyor belts have poor adaptability to complex scenarios, cannot meet real-time and lightweight requirements, and perform poorly when dealing with foreign objects with large size differences. In order to solve the above problems, a Stair-YOLOv7-tiny model is proposed based on the lightweight YOLOv7-tiny model for improvement, and applied to the detection of foreign objects in coal mine underground conveyor belts. This model adds feature concatenation units to the efficient layer aggregation network (ELAN) module to form a Stair-ELAN module. The model fuses low dimensional features from different levels with high-dimensional features, strengthens the direct connection between feature levels, enhances information capture capabilities, and strengthens the model's adaptability to objects of different scales and complex scenes. The introduction of Stair-head feature fusion (Stair-fusion) for detection heads forms a Stair-head module. The model enhances the feature expression capability of medium and low resolution detection heads by fusing detection head features of different resolutions layer by layer, achieving complementary feature information. The experimental results show that the Stair-YOLOv7 tiny model has better detection performance than CBAM-YOLOv5, YOLOv7 tiny, and its lightweight model on the open-source dataset CUMT BelT for conveyor belt foreign objects. The accuracy, average precision, recall, and precision are 98.5%, 81.0%, 82.2%, and 88.4%, respectively, and the detection speed is 192.3 frames per second. In the video analysis of conveyor belt monitoring in a certain mine, the Stair-YOLOv7-tiny model does not have any missed or false detection, achieving accurate detection of foreign objects in the conveyor belt. -
表 1 不同模型性能对比结果
Table 1. Performance comparison results of different models
模型 精确率/% 召回率/% 平均精度均值/% 准确率/% 浮点运算数/109 单帧推理时间/ms 检测速度/(帧·s−1) CBAM−YOLOv5 87.8 80.3 78.2 94.7 113.2 16.2 61.7 YOLOv7−tiny 85.3 80.1 76.5 94.2 13.0 4.8 208.3 YOLOv7−tiny−Ghost 87.5 79.5 78.0 95.4 10.4 5.9 169.5 YOLOv7−tiny−MobileNetv2 85.1 79.0 76.3 93.1 15.2 14.3 69.9 YOLOv7−tiny−ShuffleNetv2 86.1 74.4 75.2 91.6 9.1 7.0 142.9 Stair−YOLOv7−tiny 88.4 82.2 81.0 98.5 19.3 5.2 192.3 表 2 消融实验对比结果
Table 2. Comparison results of ablation experiment
Stair−ELAN Stair−head 精确率/% 召回率/% 平均精度均值/% 准确率/% 单帧推理时间/ms 检测速度/(帧·s−1) × × 85.3 80.1 76.5 94.2 4.8 208.3 √ × 86.5 81.4 77.8 95.3 4.9 204.1 × √ 87.1 82.0 79.3 97.1 5.1 196.1 √ √ 88.4 82.2 81.0 98.5 5.2 192.3 -
[1] YAN Pengcheng,SUN Quansheng,YIN Nini,et al. Detection of coal and gangue based on improved YOLOv5.1 which embedded scSE module[J]. Measurement,2022,188. DOI: 10.1016/j.measurement.2021.110530. [2] WANG Yong,JIANG Zhipeng,WANG Yihan,et al. Intelligent detection of foreign objects over coal flow based on improved GANomaly[J]. Journal of Intelligent & Fuzzy Systems,2024,46(3):5841-5851. [3] WANG Xi,GUO Yongcun,WANG Shuang,et al. Rapid detection of incomplete coal and gangue based on improved PSPNet[J]. Measurement,2022,201. DOI: 10.1016/j.measurement.2022.111646. [4] DOU Dongyang,WU Wenze,YANG Jianguo,et al. Classification of coal and gangue under multiple surface conditions via machine vision and relief-SVM[J]. Powder Technology,2019,356:1024-1028. doi: 10.1016/j.powtec.2019.09.007 [5] 王燕,郭潇樯,刘新华. 带式输送机大块异物视觉检测系统设计[J]. 机械科学与技术,2021,40(12):1939-1943.WANG Yan,GUO Xiaoqiang,LIU Xinhua. Design of visual detection system for large foreign body in belt conveyor[J]. Mechanical Science and Technology for Aerospace Engineering,2021,40(12):1939-1943. [6] 程健,王东伟,杨凌凯,等. 一种改进的高斯混合模型煤矸石视频检测方法[J]. 中南大学学报(自然科学版),2018,49(1):118-123. doi: 10.11817/j.issn.1672-7207.2018.01.016CHENG Jian,WANG Dongwei,YANG Lingkai,et al. An improved Gaussian mixture model for coal gangue video detection[J]. Journal of Central South University (Science and Technology),2018,49(1):118-123. doi: 10.11817/j.issn.1672-7207.2018.01.016 [7] PU Yuanyuan,APEL D B,SZMIGIEL A,et al. Image recognition of coal and coal gangue using a convolutional neural network and transfer learning[J]. Energies,2019,12(9). DOI: 10.3390/en12091735. [8] 程德强,徐进洋,寇旗旗,等. 融合残差信息轻量级网络的运煤皮带异物分类[J]. 煤炭学报,2022,47(3):1361-1369.CHENG Deqiang,XU Jinyang,KOU Qiqi,et al. Lightweight network based on residual information for foreign body classification on coal conveyor belt[J]. Journal of China Coal Society,2022,47(3):1361-1369. [9] 曹正远,蒋伟,方成辉. 基于双注意力生成对抗网络的煤流异物智能检测方法[J]. 工矿自动化,2023,49(12):56-62.CAO Zhengyuan,JIANG Wei,FANG Chenghui. Intelligent detection method for coal flow foreign objects based on dual attention generative adversarial network[J]. Journal of Mine Automation,2023,49(12):56-62. [10] 杨建辉,黄子洋,汪梅,等. 机器视觉灰度化金字塔卷积模型的煤流异物识别[J]. 煤炭科学技术,2022,50(11):194-201.YANG Jianhui,HUANG Ziyang,WANG Mei,et al. Recognition of unwanted objects in coal flow based on gray pyramid convolution model of machine vision[J]. Coal Science and Technology,2022,50(11):194-201. [11] 薛旭升,杨星云,齐广浩,等. 煤矿带式输送机分拣机器人异物识别与定位系统设计[J]. 工矿自动化,2022,48(12):33-41.XUE Xusheng,YANG Xingyun,QI Guanghao,et al. Design of foreign object recognition and positioning system for sorting robot of coal mine belt conveyor[J]. Journal of Mine Automation,2022,48(12):33-41. [12] 任志玲,朱彦存. 改进CenterNet算法的煤矿皮带运输异物识别研究[J]. 控制工程,2023,30(4):703-711.REN Zhiling,ZHU Yancun. Research on foreign object detection of coal mine belt transportation with improved CenterNet algorithm[J]. Control Engineering of China,2023,30(4):703-711. [13] ZHANG Mengchao,CAO Yueshuai,JIANG Kai,et al. Proactive measures to prevent conveyor belt failures:deep learning-based faster foreign object detection[J]. Engineering Failure Analysis,2022,141. DOI: 10.1016/j.engfailanal.2022.106653. [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] 高涵,赵培培,于正,等. 基于特征增强与Transformer的煤矿输送带异物检测[J]. 煤炭科学技术,2024,52(7):199-208. doi: 10.12438/cst.2023-1336GAO Han,ZHAO Peipei,YU Zheng,et al. Coal mine conveyor belt foreign object detection based on feature enhancement and Transformer[J]. Coal Science and Technology,2024,52(7):199-208. doi: 10.12438/cst.2023-1336 [16] WANG C Y,BOCHKOVSKIY A,LIAO H Y M. YOLOv7:trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]. IEEE/CVF Conference on Computer Vision and Pattern Recognition,Vancouver,2023:7464-7475. [17] 唐俊,李敬兆,石晴,等. 基于Faster−YOLOv7的带式输送机异物实时检测[J]. 工矿自动化,2023,49(11):46-52,66.TANG Jun,LI Jingzhao,SHI Qing,et al. Real time detection of foreign objects in belt conveyors based on Faster-YOLOv7[J]. Journal of Mine Automation,2023,49(11):46-52,66. [18] 付翔,秦一凡,李浩杰,等. 新一代智能煤矿人工智能赋能技术研究综述[J]. 工矿自动化,2023,49(9):122-131,139.FU Xiang,QIN Yifan,LI Haojie,et al. Summary of research on artificial intelligence empowerment technology for new generation intelligent coal mine[J]. Journal of Mine Automation,2023,49(9):122-131,139. [19] ZHANG Xindong,ZENG Hui,GUO Shi,et al. Efficient long-range attention network for image super-resolution[C]. European Conference on Computer Vision,Tel Aviv,2022:649-667. [20] ZHANG Bin,XIAO Deqin,LIU Junbin,et al. Pig eye area temperature extraction algorithm based on registered images[J]. Computers and Electronics in Agriculture,2024,217. DOI: 10.1016/j.compag.2023.108549. [21] JIA Kunming,NIU Qunfeng,WANG Li,et al. A new efficient multi-object detection and size calculation for blended tobacco shreds using an improved YOLOv7 network and LWC algorithm[J]. Sensors,2023,23(20). DOI: 10.3390/s23208380. [22] MA Ningning,ZHANG Xiangyu,ZHENG Haitao,et al. ShuffleNet V2:practical guidelines for efficient CNN architecture design[C]. European Conference on Computer Vision,Munich,2018:122-138.