[1]何建海,郑力新,臧佳明,等.基于改进图卷积网络和人体骨架的扶梯场景危险行为识别[J].华侨大学学报(自然科学版),2025,46(3):308-318.[doi:10.11830/ISSN.1000-5013.202409023]
　HE Jianhai,ZHENG Lixin,ZANG Jiaming,et al.Dangerous Behavior Recognition in Escalator Scene Based on Improved Graph Convolutional Network and Human Skeleton[J].Journal of Huaqiao University(Natural Science),2025,46(3):308-318.[doi:10.11830/ISSN.1000-5013.202409023]
点击复制

基于改进图卷积网络和人体骨架的扶梯场景危险行为识别()
分享到：

《华侨大学学报（自然科学版）》[ISSN:1000-5013/CN:35-1079/N]

卷:
第46卷
期数:
2025年第3期
页码:
308-318
栏目:
出版日期:
2025-05-20

文章信息/Info

Title:
Dangerous Behavior Recognition in Escalator Scene Based on Improved Graph Convolutional Network and Human Skeleton
文章编号:
1000-5013(2025)03-0308-11
作者:
何建海1;  郑力新1;  臧佳明1;  庄琼云2;  潘书万1
1. 华侨大学 工学院, 福建 泉州 362021;2. 黎明职业大学 信息与电子工程学院, 福建 泉州 362000
Author(s):
HE Jianhai1;  ZHENG Lixin1;  ZANG Jiaming1;  ZHUANG Qiongyun2;  PAN Shuwan1
1. College of Engineering, Huaqiao University, Quanzhou 362021, China; 2. College of Information and Electronic Engineering, Liming Vocational University, Quanzhou 362000, China
关键词:
扶梯;  图神经网络;  危险行为识别;  多尺度特征;  层次边缘卷积
Keywords:
escalator;  graph neural network;  dangerous behavior identification;  multi-scale feature;  hierarchical edge convolution
分类号:
TP391.4;TU229
DOI:
10.11830/ISSN.1000-5013.202409023
文献标志码:
A
摘要:
为了使图神经网络从前后相邻帧中获取缺失的人体骨架信息,解决自动扶梯狭长环境的遮挡问题和相似人体骨架动作准确识别问题,提出一种注意力引导的多尺度层次边缘聚合时序图卷积网络(AMHGCN)。首先,对时序卷积网络加入不同扩张率的多尺度特征,延展出的7个分支可增强网络对时间域的特征提取能力;其次,在多尺度特征时序卷积网络后面加入层次边缘卷积,使局部特征向全局特征扩张;最后,在每个时空图卷积块中,加入空间通道注意力机制,强化网络对空间、通道信息的处理,使AMHGCN在分类过程中更加关注不同行为的细节特征,提高分类的准确率。在NTU RGB+D数据集和扶梯危险行为数据集上,对AMHGCN进行评估。结果表明:相较于基线方法STGCN++,AMHGCN在NTU RGB+D数据集和扶梯危险行为数据集上的识别准确率均有较大的提高。
Abstract:
To address the occlusion problem in the narrow environment of escalators and the accurate recognition of similar human skeleton actions, a novel method called attention-guided multi-scale hierarchical edge aggregation sequential graph convolutional network(AMHGCN)is proposed to enable the graph neural network to capture missing human skeleton information from adjacent frames. Firstly, multi-scale features with different dilation rates are added to the temporal convolutional network, the extended seven branches can enhance the network’s ability to extract features in the time domain. Secondly, hierarchical edge convolution is added after the multi-scale feature temporal convolutional network to expand local features to global features. Finally, a spatial channel attention mechanism is incorporated into each spatiotemporal graph convolutional block to strengthen the network’s processing of spatial and channel information, making AMHGCN pays more atten-tion to the detailed features of different behaviors in the classification process and improves the classification accuracy. The evaluation of AMHGCN is conducted on the NTU RGB+D dataset and the escalator dangerous behavior dataset. The results show that compared to the baseline method STGCN++, AMHGCN achieves a significant improment in recognition accuracy on both the NTU RGB+D dataset and the escalator dangerous behavior dataset.

参考文献/References:

[1] 舒文华,欧阳惠卿.自动扶梯乘客行为智能感知和自主安全管理技术标准探讨[J].质量与标准化,2021,354(10):39-43.DOI:10.3969/j.issn.2095-0918.2021.10.015.
[2] 蒋儒浩.自动扶梯综合性能检测仪研制[D].合肥:合肥工业大学,2019.DOI:10.27101/d.cnki.ghfgu.2019.000056.
[3] 付春平.自动扶梯几起安全事故的共性分析与探讨[J].科技与创新,2023,217(1):82-84,89.DOI:10.15913/j.cnki.kjycx.2023.01.023.
[4] 张栓柱.基于事故树的商场电梯事故分析[J].消防界(电子版),2022,8(21):21-23.DOI:10.16859/j.cnki.cn12-9204/tu.2022.21.040.
[5] 解云蕾.自动扶梯安全探讨[J].中国科技信息,2022(3):64-66.DOI:10.3969/j.issn.1001-8972.2022.03.021.
[6] 李伟达,叶靓玲,郑力新,等.面向扶梯不安全行为的改进型深度学习检测算法[J].华侨大学学报(自然科学版),2022,43(1):119-126.DOI:10.11830/ISSN.1000-5013.202105059.
[7] 叶靓玲,李伟达,郑力新,等.结合目标检测与特征匹配的多目标跟踪算法[J].华侨大学学报(自然科学版),2021,42(5):661-669.DOI:10.11830/ISSN.1000-5013.202105018.
[8] YE Liangling,LI Weida,ZHENG Lixin,et al.Lightweight and deep appearance embedding for multiple object tracking[J].IET Computer Vision,2022,16(6):489-503.DOI:10.1049/cvi2.12106.
[9] 林志鸿,郑力新,曾远跃.采用空间依赖的MTDPN扶梯危险行为的姿态估计[J].华侨大学学报(自然科学版),2023,44(6):751-758.DOI:10.11830/ISSN.1000-5013.202305020.
[10] YAN Sijie,XIONG Yuanjun,LIN Dahua.Spatial temporal graph convolutional networks for skeleton-based action recognition[C]//Proceedings of the AAAI Conference on Artificial Intelligence.New Orleans:AAAI Press,2018:7444-7452.DOI:10.48550/arXiv.1801.07455.
[11] ZHANG Xikun,XU Chang,TIAN Xinmei,et al.Graph edge convolutional neural networks for skeleton-based action recognition[J].IEEE Transactions on Neural Networks and Learning Systems,2019,31(8):3047-3060.DOI:10.48550/arXiv.1805.06184.
[12] SHI Lei,ZHANG Yifan,CHENG Jian,et al.Two-stream adaptive graph convolutional networks for skeleton-based action recognition[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Long Beach:IEEE Press,2019:12026-12035.DOI:10.48550/arXiv.1805.07694.
[13] WEN Yuhui,GAO Lin,FU Hongbo,et al.Graph CNNs with motif and variable temporal block for skeleton-based action recognition[C]//Proceedings of the AAAI Conference on Artificial Intelligence.Honolulu:AAAI Press,2019:8989-8996.DOI:10.1609/aaai.v33i01.33018989.
[14] LI Bin,LI Xi,ZHANG Zhongfei,et al.Spatio-temporal graph routing for skeleton-based action recognition[C]//Proceedings of the AAAI Conference on Artificial Intelligence.Honolulu:AAAI Press,2019:8561-8568.DOI:10.1609/aaai.v33i01.33018561.
[15] LI Maosen,CHEN Siheng,CHEN Xu,et al.Actional-structural graph convolutional networks for skeleton based action recognition[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Long Beach:IEEE Press,2019:3595-3603.DOI:10.1109/CVPR.2019.01230.
[16] SHI Lei,ZHANG Yifan,CHENG Jian,et al.Skeleton-based action recognition with directed graph neural networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Long Beach:IEEE Press,2019:7912-7921.DOI:10.1109/CVPR.2019.00810.
[17] LEE I,KIM D,KANG S,et al.Ensemble deep learning for skeleton-based action recognition using temporal sliding lstm networks[C]//Proceedings of the IEEE International Conference on Computer Vision.Venice:IEEE Press,2017:1012-1020.DOI:10.1109/ICCV.2017.115.
[18] CHEN Yuxin,ZHANG Ziqi,YUAN Chunfeng,et al.Channel-wise topology refinement graph convolution for skeleton-based action recognition[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.Montreal:IEEE Press,2021:13359-13368.DOI:10.48550/arXiv.2107.12213.
[19] 袁正中,李灿东.复杂网络控制核心的进一步分析[J].闽南师范大学学报(自然科学版),2023,36(2):27-34.DOI:10.16007/j.cnki.issn2095-7122.2023.02.008.
[20] LIU Ziyu,ZHANG Hongwen,CHEN Zhenghao,et al.Disentangling and unifying graph convolutions for skeleton-based action recognition[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Seattle:IEEE Press,2020:143-152.DOI:10.1109/CVPR42600.2020.00022.
[21] SHAHROUDY A,LIU Jun,NG T T,et al.Ntu rgb+d: A large scale dataset for 3d human activity analysis[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas:IEEE Press,2016:1010-1019.DOI:10.1109/CVPR.2016.115.
[22] DUAN Haodong,WANG Jiaqi,CHEN Kai,et al. Pyskl: Towards good practices for skeleton action recognition[C]//Proceedings of the 30th ACM International Conference on Multimedia.Nicosia:ACM Press,2022:7351-7354.DOI:10.48550/arXiv.2205.09443.
[23] SHI Lei,ZHANG Yifan,CHENG Jian,et al.Skeleton-based action recognition with multi-stream adaptive graph convolutional networks[C]//IEEE Transactions on Image Processing.[S.l.]:IEEE Press,2020:9532-9545.DOI:10.1109/TIP.2020.3028207.
[24] LIU Ziyu,ZHANG Hongwen,CHEN Zhenghao,et al.Disentangling and unifying graph convolutions for skeleton-based action recognition[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Seattle:IEEE Press,2020:143-152.DOI:10.1109/CVPR42600.2020.00022.
[25] CHEN Yuxin,ZHANG Ziqi,YUAN Chunfeng,et al.Channel-wise topology refinement graph convolution for skeleton-based action recognition[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.Montreal:IEEE Press,2021:13359-13368.DOI:10.48550/arXiv.2107.12213.

相似文献/References:

[1]李伟达,叶靓玲,郑力新,等.面向扶梯不安全行为的改进型深度学习检测算法[J].华侨大学学报(自然科学版),2022,43(1):119.[doi:10.11830/ISSN.1000-5013.202105059]
　LI Weida,YE Liangling,ZHENG Lixin,et al.Improved Deep Learning Detection Algorithm for Unsafe Escalator Behavior[J].Journal of Huaqiao University(Natural Science),2022,43(3):119.[doi:10.11830/ISSN.1000-5013.202105059]

备注/Memo

备注/Memo:
收稿日期: 2024-09-29
通信作者: 郑力新(1967-),男,教授,博士,主要从事图像分析、机器视觉和深度学习方法的研究。 E-mail:zlx@hqu.edu.cn。
基金项目: 福建省科技计划重点项目(2020Y0039); 黎明职业大学2022年度校级一般课题(自然科学类)(LZ 202211)http://hdxb.hqu.edu.cn/
更新日期/Last Update: 2025-05-20