|本期目录/Table of Contents|

[1]邹子寒,李华,陈宇飞,等.基于改进FaceNet模型的施工作业人员遮挡人脸身份识别研究*[J].中国安全生产科学技术,2025,21(5):140-146.[doi:10.11731/j.issn.1673-193x.2025.05.018]
　ZOU Zihan,LI Hua,CHEN Yufei,et al.Research on obstructed face recognition for construction workers based on improved FaceNet model[J].Journal of Safety Science and Technology,2025,21(5):140-146.[doi:10.11731/j.issn.1673-193x.2025.05.018]
点击复制

基于改进FaceNet模型的施工作业人员遮挡人脸身份识别研究*

《中国安全生产科学技术》[ISSN:1673-193X/CN:11-5335/TB]

卷:
21
期数:
2025年5期
页码:
140-146
栏目:
职业安全卫生管理与技术
出版日期:
2025-05-30

文章信息/Info

Title:
Research on obstructed face recognition for construction workers based on improved FaceNet model
文章编号:
1673-193X(2025)-05-0140-07
作者:
邹子寒; 李华; 陈宇飞; 钟兴润
（1.西安建筑科技大学 资源工程学院，陕西 西安 710055；
2.西安电子科技大学 网络空间与安全学院，陕西 西安 710126 ）
Author(s):
ZOU Zihan;  LI Hua;  CHEN Yufei;  ZHONG Xingrun
(1.School of Resources Engineering,Xi’an University of Architecture and Technology,Xi’an Shaanxi 710055,China;
2.School of Cyber Engineering,Xidian University,Xi’an Shaanxi 710126,China)
关键词:
建筑工地; 人脸识别; 遮挡; 安全管理; FaceNet; MTCNN; 特种设备
Keywords:
construction site;  facial recognition;  occlusion;  safety management;  FaceNet;  MTCNN;  special equipment
分类号:
X947
DOI:
10.11731/j.issn.1673-193x.2025.05.018
文献标志码:
A
摘要:
为提高建筑工地中特种设备操作人员在安全帽遮挡情况下身份验证的准确性，提出1种结合FaceNet和MTCNN算法的人脸识别方法。首先由MTCNN通过三级联结构有效检测并定位人脸关键点，随后由FaceNet提取遮挡情况下的人脸特征并进行身份验证。实验数据集包括LFW公开人脸数据集和工地现场采集的真实数据，并通过数据增强方法模拟遮挡和光照情况。研究结果表明：在未佩戴安全帽的LFW数据集上的精准率达到96.5%，在佩戴安全帽的工地数据集和模拟数据集上精准率为95.8%，平均精准率达到96.15%；与常用的Dlib和OpenCV人脸识别算法相比，遮挡情况下的识别精准率显著提升，且处理速度满足实时应用需求，能有效提升工地人员身份验证的准确性和可靠性。研究结果可为建筑工地安全管理提供技术参考。
Abstract:
To enhance the accuracy of identity verification for special equipment operators on construction sites under the conditions of safety helmet occlusion,a facial recognition method combining FaceNet and MTCNN algorithms was proposed.Initially,MTCNN detected and localized the facial key points effectively through a three-stage cascade structure,and the FaceNet extracted the facial features under occlusion conditions for identity verification.The experimental dataset included the LFW public face dataset and real data collected from construction sites,and the occlusion and lighting conditions were simulated through the data augmentation method.The results show that the precision rate on the LFW dataset without helmet is 96.5%,and that on the construction site dataset and simulated dataset with helmets is 95.8%,with an average precision rate of 96.15%.Compared to the commonly used Dlib and OpenCV facial recognition algorithms,the recognition precision rate under occlusion conditions has significantly improved,and the processing speed meets the requirements for real-time application,which can effectively enhance the accuracy and reliability of identity verification for construction site personnel.The research results can provide technical support for the safety management of construction sites.

参考文献/References:

［1］市场监管总局关于2022年全国特种设备安全状况的通告［J］.中国特种设备安全,2023,39(3):1-3. Notice of thestate administration for market eegulation on the safety status of special equipment in China in 2022 ［J］.China Special Equipment Safety,2023,39(3):1-3.
［2］ABID M M,MAHMOOD T,ASHRAF R,et al.Computationally intelligent real-time security surveillance system in the education sector using deep learning［J］.PloS One,2024,19(7):e0301908.
［3］MINHAS R,PEKER N Y,HAKKOZ M A,et al.Association of visual-based signals with electroencephalography patterns in enhancing the drowsiness detection in drivers with obstructive sleep apnea［J］.Sensors,2024,24(8):2625.
［4］肖森,安嘉骏,高雨,等.低本高效人脸识别系统设计的汽车记忆座椅模块开发［J］.科学技术与工程,2024,24(13):5443-5449. XIAO Sen,AN Jiajun,GAO Yu,et al.Development of automotive memory seat module with low cost and efficient facial recognition system design ［J］.Science Technology and Engineering,2024,24(13):5443-5449.
［5］张铮,芦天亮,曹金璇.基于分割和多级掩膜学习的遮挡人脸识别方法［J］.计算机科学与探索,2024,18(7):1814-1825. ZHANG Zheng,LU Tianliang,CAO Jinxuan.Obstructive facial recognition method based on segmentation and multi-level mask learning ［J］.Computer Science and Exploration,2024,18(7):1814-1825.
［6］JIANG B,LI N,CUI X,et al.Research on facial expression recognition algorithm based on lightweight transformer［J］.Information,2024,15(6):321.
［7］杨显,王志伟.工程建设中基于实名制和地理定位的人脸识别移动考勤设计与实现［J］.施工技术(中英文),2022,51(8):127-130. YANG Xian,WANG Zhiwei.Design and implementation of facial recognition mobile attendance based on real name system and geographic positioning in engineering construction ［J］.Construction Technology (Chinese and English),2022,51(8):127-130.
［8］陈润彬.基于图像识别的电力工人智能安防安全检测系统研究［D］.广州：广东工业大学,2021.
［9］范冰倩,董秉聿,王彪,等.基于深度学习的地铁施工作业人员不安全行为识别与应用［J］.中国安全科学学报,2023,33(1):41-47. FAN Bingqian,DONG Bingyu,WANG Biao,et al.Identification and application of unsafe behaviors of subway constructionworkers based on deep learning［J］.China Safety Science Journal,2023,33(1):41-47.
［10］SCHROFF F,KALENICHENKO D,PHILBIN J.FaceNet:a unified embedding for face recognition and clustering［C］//Proceedings of the IEEE conference on computer vision and pattern recognition,2015:815-823.
［11］ZHANG K,ZHANG Z,LI Z,et al.Joint face detection and alignment using multitask cascaded convolutional networks［J］.IEEE signal processing letters,2016,23(10):1499-1503.
［12］LONG J,SHELHAMER E,DARRELL T.Fully convolutional networks for semantic segmentation［C］//Proceedings of the IEEE conference on computer vision and pattern recognition.2015:3431-3440.
［13］WILLIAM I,RACHMAWANTO E H,SANTOSO H A,et al.Face recognition using facenet (survey,performance test,and comparison)［C］//2019 fourth international conference on informatics and computing (ICIC),IEEE,2019:1-6.
［14］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 the application of YOLOv4 object detection algorithm in monitoring mask wearing by coal mine workers ［J］.Journal of Safety Science and Technology,2021,17(10):66-71.
［15］HUANG G B,MATTAR M,BERG T,et al.Labeled faces in the wild:a database forstudying face recognition in unconstrained environments［C］//Workshop on faces in“Real-Life”Images:detection,alignment,and recognition,2008.

相似文献/References:

[1]李清江.建筑工地消防安全问题分析与对策[J].中国安全生产科学技术,2012,8(7):229.
　LI Qing jiang.Construction site fire safety analysis and countermeasures[J].Journal of Safety Science and Technology,2012,8(5):229.
[2]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.[doi:10.11731/j.issn.1673-193x.2021.10.010]
　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(5):66.[doi:10.11731/j.issn.1673-193x.2021.10.010]

备注/Memo

备注/Memo:
收稿日期： 2024-08-15
* 基金项目： 陕西省建设厅科技发展计划项目(2020-K32)；西安建筑科大工程技术有限公司2023年度科研项目(XAJD-YF23N010)
作者简介： 邹子寒，硕士研究生，主要研究方向为建筑安全、应急管理。
通信作者： 李华，博士，副教授，主要研究方向为企业风险评估与安全管理、建筑安全监测与监控、公共安全与应急管理等。
更新日期/Last Update: 2025-05-26