Intelligent detection method of working personnel wearing safety helmets in underground mine
-
摘要: 基于视觉图像方法是目前矿井人员佩戴安全帽智能检测的热点,但现有方法所用的地下矿山数据较少,安全帽特征分类不够精确。通过采集地下矿山采场、井巷等实际生产场景的图像,构建了矿山安全帽佩戴数据集——MHWD,并将安全帽佩戴情况进一步细分为正确佩戴、不规范佩戴和未佩戴3类。采用YOLOX算法进行人员佩戴安全帽检测,为了增强YOLOX提取全局特征的能力,引入注意力机制,即在YOLOX骨干网的空间金字塔池化瓶颈层嵌入有效通道注意力模块,在路径聚合特征金字塔网络每个上采样和下采样后添加卷积块注意力模块,由此构建了YOLOX−A模型。采用MHWD训练YOLOX−A模型并进行验证,结果表明,针对照度低、模糊、有人员遮挡的矿井图像,YOLOX−A模型能够准确识别人员佩戴安全帽情况,对不规范佩戴、正确佩戴和未佩戴安全帽3种分类结果的F1分数分别为0.86,0.92,0.89,平均精度分别为93.16%,95.76%,91.69%,平均精度均值为93.54%,整体F1分数较YOLOX模型提升4%,检测精度高于主流目标检测模型EfficientDet,YOLOv3,YOLOv4,YOLOv5,YOLOX。Abstract: Visual image-based methods are currently a hot topic in intelligent detection of mine personnel wearing safety helmets. However, existing methods use limited underground mining data and the classification of safety helmet features is not accurate enough. By collecting images of actual production scenes such as underground mining sites and roadways, a mining helmet wearing dataset (MHWD) is constructed. The helmet wearing situation is further divided into three categories: correct wearing, non-standard wearing, and non wearing. YOLOX algorithm is used to detect personnel wearing helmets. In order to enhance YOLOX's capability to extract global features, the attention mechanism is introduced. The effective channel attention module is embedded in the spatial pyramid pooling bottleneck layer of YOLOX's backbone network. The convolutional block attention module is added after each upsampling and downsampling of the path aggregation feature pyramid network, thus the YOLOX-A model is built. By using MHWD, the YOLOX-A model is trained and validated. The results show that the YOLOX-A model can accurately identify the wearing of safety helmets by personnel in mine images with low illumination, blurriness, and personnel obstruction. The F1 scores for the classification results of non-standard wearing, correct wearing, and non wearing safety helmets are 0.86, 0.92, and 0.89, respectively. The average precision is 93.16%, 95.76%, and 91.69%. The average precision mean is 93.54%. The overall F1 score is 4% higher than the YOLOX model. The detection precision is higher than the mainstream target detection models EfficientDet, YOLOv3, YOLOv4, YOLOv5 and YOLOX.
-
表 1 不同目标检测模型在MHWD上的检测指标
Table 1. Detection indexes of different target detection models on MHWD
% 模型 AP mAP IrregularWearing WithHelmet Person EfficientDet 83.08 90.80 37.53 70.47 YOLOv3 74.87 89.21 79.12 81.06 YOLOv4 75.36 89.23 80.63 81.74 YOLOv5 77.3 90.53 87.22 85.02 YOLOX 91.67 95.48 92.28 93.15 YOLOX−A 93.16 95.76 91.69 93.54 表 2 消融实验结果
Table 2. Ablation experiment results
YOLOX CBAM ECA mAP/% √ 93.15 √ √ 93.27 √ √ √ 93.54 -
[1] 李超. 现代化矿山救护技术装备问题分析[J]. 中国金属通报,2021(11):116-117.LI Chao. Analysis of modern mine rescue technology and equipment[J]. China Metal Bulletin,2021(11):116-117. [2] 陈杰. 智慧矿山安全防控多系统井下融合与应急联动技术研究[J]. 煤矿安全,2022,53(5):99-105.CHEN Jie. Research on multi-system underground integration and emergency linkage technology for smart mine safety prevention and control[J]. Safety in Coal Mines,2022,53(5):99-105. [3] 张立艺,武文红,牛恒茂,等. 深度学习中的安全帽检测算法应用研究综述[J]. 计算机工程与应用,2022,58(16):1-17. doi: 10.3778/j.issn.1002-8331.2203-0580ZHANG Liyi,WU Wenhong,NIU Hengmao,et al. Summary of application research on helmet detection algorithm based on deep learning[J]. Computer Engineering and Applications,2022,58(16):1-17. doi: 10.3778/j.issn.1002-8331.2203-0580 [4] 孙国栋,李超,张航. 融合自注意力机制的安全帽佩戴检测方法[J]. 计算机工程与应用,2022,58(20):300-304. doi: 10.3778/j.issn.1002-8331.2103-0372SUN Guodong,LI Chao,ZHANG Hang. Safety helmet wearing detection method fused with self-attention mechanism[J]. Computer Engineering and Applications,2022,58(20):300-304. doi: 10.3778/j.issn.1002-8331.2103-0372 [5] 李晓宇,陈伟,杨维,等. 基于超像素特征与SVM分类的人员安全帽分割方法[J]. 煤炭学报,2021,46(6):2009-2022.LI Xiaoyu,CHEN Wei,YANG Wei,et al. Segmentation method for personnel safety helmet based on super-pixel features and SVM classification[J]. Journal of China Coal Society,2021,46(6):2009-2022. [6] 毕林,谢伟,崔君. 基于卷积神经网络的矿工安全帽佩戴识别研究[J]. 黄金科学技术,2017,25(4):73-80. doi: 10.11872/j.issn.1005-2518.2017.04.073BI Lin,XIE Wei,CUI Jun. Identification research on the miner's safety helmet wear based on convolutional neural network[J]. Gold Science and Technology,2017,25(4):73-80. doi: 10.11872/j.issn.1005-2518.2017.04.073 [7] 仝泽友,冯仕民,侯晓晴,等. 基于安全帽佩戴检测的矿山人员违规行为研究[J]. 电子科技,2019,32(9):26-31. doi: 10.16180/j.cnki.issn1007-7820.2019.09.006TONG Zeyou,FENG Shimin,HOU Xiaoqing,et al. Recognition of underground miners' rule-violated behavior based on safety helmet detection[J]. Electronic Science and Technology,2019,32(9):26-31. doi: 10.16180/j.cnki.issn1007-7820.2019.09.006 [8] REDMON J, FARHADI A. Yolov3: an incremental improvement[EB/OL]. [2022-09-03]. https://arxiv.org/abs/1804.02767. [9] BOCHKOVSKI A, WANG C Y, LIAO H Y M. Yolov4: optimal speed and accuracy of object detection[EB/OL]. [2022-09-03]. https://arxiv.org/abs/2004.10934. [10] GE Zheng, LIU Songtao, WANG Feng, et al. Yolox: exceeding YOLO series in 2021[EB/OL]. [2022-09-03]. https://arxiv.org/abs/2107.08430. [11] JAMTSHO Y,RIYAMONGKOL P,WARANUSAST R. Real-time license plate detection for non-helmeted motorcyclist using YOLO[J]. ICT Express,2021,7(1):104-109. doi: 10.1016/j.icte.2020.07.008 [12] SRIDHAR P, JAGADEESWARI M, SRI S H, et al. Helmet violation detection using YOLO v2 deep learning framework[C]. The 6th International Conference on Trends in Electronics and Informatics, Tirunelveli, 2022: 1207-1212. [13] CHEN Meixi, KONG Rong, ZHU Jianming, et al. Application research of safety helmet detection based on low computing power platform using YOLO v5[C]. International Conference on Adaptive and Intelligent Systems, Suzhou, 2022: 107-117. [14] HE Zhiwei, WU Fan, GAO Mingyu, et al. Helmet detection based on improved YOLO v3 deep model[C]. IEEE 16th International Conference on Networking, Sensing and Control, Alberta, 2019: 363-368. [15] XIE Wenqin,XIE Lei,ZHANG Linzhi,et al. Toward efficient safety helmet detection based on Yolov5 with hierarchical positive sample selection and box density filtering[J]. IEEE Transactions on Instrumentation and Measurement,2022,71:1-14. [16] SHIRMOHAMMADI S,FERRERO A. Camera as the instrument:the rising trend of vision based measurement[J]. IEEE Instrumentation & Measurement Magazine,2014,17(3):41-47. [17] WANG C Y, LIAO H Y M, WU Y H, et al. CSPNet: a new backbone that can enhance learning capability of CNN[C]. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, 2020: 390-391. [18] LIN T Y, DOLLAR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Honolulu, 2017: 2117-2125. [19] HE Kaiming,ZHANG Xiayu,REN Shaoqing,et al. Spatial pyramid pooling in deep convolutional networks for visual recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,37(9):1904-1916. doi: 10.1109/TPAMI.2015.2389824 [20] WANG Qilong, WU Banggu, ZHU Pengfei, et al. ECA-Net: efficient channel attention for deep convolutional neural networks[C]. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Seattle, 2020: 11534-11542. [21] WOO S, PARK J, LEE J Y, et al. CBAM: Convolutional block attention module[C]. Proceedings of the European Conference on Computer Vision, Munich, 2018: 3-19. -