隧道作业场景下基于特征融合的人员面罩检测模型

doi:10.16265/j.cnki.issn1003-3033.2026.01.1034

摘要/Abstract

摘要：

为提高隧道作业场景下工人佩戴面罩的检测效率,提出基于特征融合的隧道作业人员面罩检测模型。首先,选取高质量查询图像并确定图像候选库,采用图像检索方法获取样本,度量查询图像与图像候选库的相似性,迭代扩充数据集;然后,从图像中提取方向梯度直方图(HOG)和Fisher特征,引入蚁狮优化算法(ALO),计算2种特征的最佳权重组合,并融合2种特征;最后,基于融合后的特征,利用支持向量机(SVM)训练面罩佩戴检测模型,在自建数据集上进行效果评估。结果表明:所提出的模型能够完成隧道作业场景下的面罩佩戴检测任务,特征融合能增强对图像的描述,提高模型的检测准确率,相比于单HOG特征和Fisher特征,准确率分别提升6%和14%,可满足隧道作业场景的工人面罩佩戴检测准确率要求。

关键词: 隧道作业场景, 特征融合, 面罩佩戴检测, 自建数据集, 支持向量机(SVM)

Abstract:

In order to improve the efficiency of facepiece-wearing detection for tunnel operation workers, a feature fusion-based facepiece detection model was proposed. First, high quality query images were selected, and an image gallery was established. An image retrieval method was adopted to obtain samples and measure the similarity between query and gallery images, thereby iteratively expanding the dataset scale. Then, Histogram of Oriented Gradients (HOG) and Fisher features were extracted from the images. The Ant Lion Optimizer (ALO) was introduced to compute the optimal weight combination for the two types of features, which were subsequently fused. Finally, based on the fused features, a SVM was utilized to train a facepiece detection model, and experimental evaluations were conducted on the self-constructed dataset. The results indicate that the proposed model effectively accomplishes the task of facepiece-wearing detection in tunnel operation scenarios. Feature fusion enhances the image description and improves the detection accuracy of the model. Compared to using only HOG features or Fisher features, the accuracy is increased by 6% and 14%, respectively. The model meets the accuracy requirements for facepiece-wearing detection of workers in tunnel construction environments.

Key words: tunnel operation scenarios, feature fusion, facepiece-wearing detection, self-constructed dataset, support vector machine(SVM)

中图分类号:

X924.4安全控制技术

柯彬彬, 孙辰晨. 隧道作业场景下基于特征融合的人员面罩检测模型[J]. 中国安全科学学报, 2026, 36(1): 267-274.

KE Binbin, SUN Chenchen. Facepiece detection model based on feature fusion for personnel in tunnel operation scenarios[J]. China Safety Science Journal, 2026, 36(1): 267-274.

图/表 10

图1

图2

图3

图4

图5

表1

表2

表3

图6

图7

参考文献 22

[1]	中华人民共和国国家卫生健康委员会. 国家卫生健康委员会2025年4月28日新闻发布会文字实录[EB/OL]. (2025-04-28). https://www.nhc.gov.cn/xcs/c100122/202504/33267e386e5e41498cca637aed41c6c5.shtml.
[2]	马世伟, 孔宪会, 阮志刚, 等. 基于事故树分析法的铁路隧道施工新发尘肺病关键控制环节研究[J]. 铁路节能环保与安全卫生, 2020, 10(6): 34-37.
	MA Shiwei, KONG Xianhui, RUAN Zhigang, et al. Study on key control links of new pneumoconiosis in railway tunnel construction based on fault tree analysis[J]. Railway Energy Saving & Environmental Protection & Occupational Safety and Health, 2020, 10(6): 34-37.
[3]	马世伟, 孔宪会, 郭海峰, 等. 高原长大隧道尘肺病防治技术发展与应用研究[J]. 铁路节能环保与安全卫生, 2021, 11 (2): 41-44.
	MA Shiwei, KONG Xianhui, GUO Haifeng, et al. Research on development and application of prevention and control technology of pneumoconiosis in long tunnel on plateau[J]. Railway Energy Saving & Environmental Protection & Occupational Safety and Health, 2021, 11 (2): 41-44.
[4]	郭春, 宋骏修, 王欣, 等. 矿山法施工隧道粉尘控制技术研究现状及进展[J]. 隧道建设:中英文, 2020, 40(增1): 68-74.
	GUO Chun, SONG Junxiu, WANG Xin, et al. Research status and development of dust control technology in tunnel constructed by mining method[J]. Tunnel Construction, 2020, 40(S1):68-74.
[5]	崔铁军, 王凌霄. YOLOv4目标检测算法在煤矿工人口罩佩戴监测工作中的应用研究[J]. 中国安全生产科学技术, 2021, 17 (10): 66-71.
	CUI Tiejun, WANG Lingxiao. Research on application of YOLOv4 object detection algorithm in monitoring on masks wearing of coal miners[J]. Journal of Safety Science and Technology, 2021, 17 (10): 66-71.
[6]	李华, 吴立舟, 钟兴润, 等. 基于计算机视觉的人员疲劳状态与不安全行为识别[J]. 中国安全科学学报, 2025, 35(3): 28-35. doi: 10.16265/j.cnki.issn1003-3033.2025.03.0749
	LI Hua, WU Lizhou, ZHONG Xingrun, et al. Recognition of personnel fatigue state and unsafe behavior based on computer vision[J]. China Safety Science Journal, 2025, 35(3): 28-35. doi: 10.16265/j.cnki.issn1003-3033.2025.03.0749
[7]	赵树磊, 孙兵, 陈稳干, 等. 超高海拔隧道施工人员生理指标变化及劳动强度研究[J]. 中国安全科学学报, 2024, 34(4): 239-46. doi: 10.16265/j.cnki.issn1003-3033.2024.04.1748
	ZHAO Shulei, SUN Bing, CHEN Wen'gan, et al. Physiological indicators and labor intensity of tunnel construction workers at ultra-high altitude[J]. China Safety Science Journal, 2024, 34(4): 239-246. doi: 10.16265/j.cnki.issn1003-3033.2024.04.1748
[8]	RAFIDISON M A, RAKOTOMIHAMINA A H, RAFANANTENANA S H J, et al. Neural networks contribution in face mask detection to reduce the spread of COVID-19[J]. Multimedia Tools and Applications, 2023, 82(21): 1-23. doi: 10.1007/s11042-022-12047-3
[9]	ABOU CHAAYA J, ZARAKET B, HARB H, et al. Convolutional neural network-based architecture for detecting face mask in crowded areas[C]. Proceeding in 22^nd IEEE Statistical Signal Processing Workshop (SSP), 2023: 408-412.
[10]	CRESPO F, CRESPO A, SIERRA-MARTíNEZ L M, et al. A computer vision model to identify the incorrect use of face masks for COVID-19 awareness[J]. Applied Sciences, 2022, 12(14): DOI:10.3390/app12146924.
[11]	LOEY M, MANOGARAN G, TAHA M H N, et al. A hybrid deep transfer learning model with machine learning methods for face mask detection in the era of the COVID-19 pandemic[J]. Measurement, 2021, 167:DOI: 10.1016/j.measurement.2020.108288.
[12]	KUMAR B A, BANSAL M. Face mask detection on photo and real-time video images using Caffe-MobileNetV2 transfer learning[J]. Applied Sciences:Switzerland, 2023, 13(2):DOI:10.3390/app13020935.
[13]	SHAO Yanhua, NING Jiajia, SHAO Huicao, et al. Lightweight face mask detection algorithm with attention mechanism[J]. Engineering Applications of Artificial Intelligence, 2024, 137:DOI:10.1016/j.engappai.2024.109077.
[14]	高民, 陈高华, 古佳欣, 等. FLM-YOLOv8:一种轻量级的口罩佩戴检测算法[J]. 计算机工程与应用, 2024, 60(17): 203-215. doi: 10.3778/j.issn.1002-8331.2402-0226
	GAO Min, CHEN Gaohua, GU Jiaxin, et al. FLM-YOLOv8:lightweight mask wearing detection algorithm[J]. Computer Engineering and Applications, 2024, 60(17): 203-215. doi: 10.3778/j.issn.1002-8331.2402-0226
[15]	NAGRATH P, JAIN R, MADAN A, et al. SSDMNV2:a real time DNN-based face mask detection system using single shot multibox detector and MobileNetV2[J]. Sustainable Cities and Society, 2021, 66:DOI:10.1016/j.scs.2020.102692.
[16]	WEN Peng, YUAN Zhengyi, ZHANG Junhu, et al. MF-YOLO: mask wearing detection algorithm for dense environments[J]. IEEE Access, 2025, 13:DOI: 10.1109/ACCESS.2025.3551892.
[17]	SINGH D, SINGH B. Hybridization of feature selection and feature weighting for high dimensional data[J]. Applied Intelligence, 2019, 49(4): 1580-1596. doi: 10.1007/s10489-018-1348-2
[18]	CORTES C, VAPNIK V. Support-vector networks[J]. Machine Learning, 1995, 20(3): 273-297. doi: 10.1023/A:1022627411411
[19]	DALAI N, TRIGGS B. Histograms of oriented gradients for human detection[C]. 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05), 2005: 886-893.
[20]	LOWE D G. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004, 60(2): 91-110. doi: 10.1023/B:VISI.0000029664.99615.94
[21]	REYNOLDS D A, QUATIERI T F, DUNN R B. Speaker verification using adapted gaussian mixture models[J]. Digital Signal Processing, 2000, 10(1/2/3): 19-41. doi: 10.1006/dspr.1999.0361
[22]	PERRONNIN F, DANCE C. Fisher kernels on visual vocabularies for image categorization[C]. 2007 IEEE Conference on Computer Vision and Pattern Recognition, 2007: 1-8.

特征	类型
面罩佩戴状态	佩戴面罩、未佩戴面罩
背景	昏暗自然光、昏暗人造光、明亮人造光
尺度	近距离、远距离
角度	正面、侧面