文献类型：期刊文献

中文题名：盾构隧道施工场景下的作业人员非正面人脸识别方法

英文题名：Non-front View Face Recognition Method for Construction Workers in Shield Tunneling Scenarios

作者：王军[1,2];马浴阳[2];林瀚[3];陈烨斌[3];谭帅[3]

机构：[1]东南大学网络空间安全学院,南京211189;[2]中铁十四局集团有限公司,济南250101;[3]华东理工大学能源化工过程智能制造教育部重点实验室,上海200237

年份：2026

卷号：29

期号：6

起止页码：109

中文期刊名：城市轨道交通研究

外文期刊名：Urban Mass Transit

收录：;北大核心:【北大核心2023】；

基金：国家自然科学基金项目(62273147);中铁十四局集团科技研发计划课题(913700001630559891202215);第六届中国科协青托项目(2020-2022QNRC001)。

语种：中文

中文关键词：城市轨道交通;盾构隧道;非正面人脸识别;多姿态矫正

外文关键词：urban rail transit;shield tunnel;non-front view face recognition;multi-posture correction

摘要：[目的]随着我国基建事业的快速发展,以盾构施工为代表的建筑工程领域的安全生产问题日渐突出。目前,施工现场监控多依赖人工监督,监控效率低且难以保障施工作业人员的生命安全。针对施工现场人员安全帽违规佩戴识别中,人脸识别算法常面临的非正面人脸导致身份核验困难问题,开展相关研究。[方法]首先进行数据预处理,针对盾构机掘进过程中振动引发的图像噪声干扰,采用NLM(非局部均值)图像去噪算法进行图像去噪处理。针对施工场景下的非正面人脸特征缺失问题,对传统基于DREAM(深度残差等变映射)的人脸识别算法进行改进,通过提取人脸姿态与人脸特征之间的相互关系,多维度考量人脸姿态对于特征矫正的作用,提出一种基于多姿态矫正的非正面人脸识别算法。[结果及结论]所提算法能够根据人脸姿态角对所提取到的人脸特征进行矫正,进而实现输出人脸特征与姿态无关,解决了人脸图像数据存在多姿态的问题,提高了算法模型对盾构隧道施工场景下的非正面人脸识别准确率。
[Objective]With the rapid development of infrastructure industry in China,safety production issues in the construction field,represented by shield tunneling,are becoming increasingly prominent.Currently,construction site monitoring relies heavily on manual supervision,leading to low efficiency and difficulty in ensuring construction workers’safety.Regarding the identity verification difficulties in facial recognition algorithms for non-front view faces caused by on-site helmet wearing violations,corresponding researches are conducted.[Method]First,data preprocessing is performed.To address the image noise interference caused by vibration during shield tunneling process,the NLM(non-local means)algorithm is adopted for image denoising.To address the missing features of non-front view faces in construction scenarios,the traditional face recognition algorithm based on DREAM(deep residual equivariant mapping)is improved.By extracting the relationship between the face poses and face features,and considering the effect of face poses on feature correction from multiple dimensions,a non-front view face recognition algorithm based on multi-pose correction is proposed.[Result&Conclusion]The proposed algorithm can correct the extracted face features according to the face pose angles,thereby realizing the output face features pose-invariant.This solves the problem of multipose face image data and improves the accuracy of non-front view face recognition in shield tunnel construction scenarios.