首页 > 论文 > 激光与光电子学进展 > 57卷 > 2期(pp:21009--1)

基于卷积神经网络的雷达人体动作与身份多任务识别

Human Activity and IdentityMulti-Task Recognition Based on Convolutional Neural Network Using Doppler Radar

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

为弥补单任务识别未充分利用相关任务监督信息的缺陷,提出了一种基于卷积神经网络的多任务识别模型。该模型引入注意力机制,对任务共享层的特征进行重校正,并结合多尺度结构进行特征融合,最后在任务特定层上进行多任务识别。针对共享特征空间内类分布不紧凑导致的模型泛化性能降低问题,本文在模型中引入中心损失函数与均方误差损失函数,与传统的交叉熵损失函数相结合,共同优化模型。实验结果表明:所提模型在人体6个动作类别和15个身份类别上的最高识别准确率分别可达100%和99.93%,两种任务上识别的总准确率可达99.93%,均优于任务独立识别时的各项准确率,说明所提模型能更有效地同时完成人体动作及身份识别任务。

Abstract

A multitask recognition model based on convolutional neural network is proposed to avoid single task recognition ignoring supervision information of related tasks. The proposed model introduces an attention mechanism to perform feature recalibration of the task shared layer and combines the multiscale structure for feature fusion. Finally, multi-task recognition is performed on the task-specific layers. Center loss and mean square error loss functions are employed together with the traditional cross entropy loss function to solve the generalization degradation problem caused by uncompact class distribution in the shared feature space. Experimental results on 6 human activities and 15 identities show that the model can achieve the maximum recognition accuracies of 100% and 99.93% on each task, respectively, and the multitask accuracy is up to 99.93%. The results are better than those obtained by the single task models. This shows that the model can simultaneously perform human activity and identity recognition more effectively.

Newport宣传-MKS新实验室计划
补充资料

中图分类号:TP391.4

DOI:10.3788/LOP57.021009

所属栏目:图像处理

基金项目:国家自然科学基金;

收稿日期:2019-06-10

修改稿日期:2019-07-01

网络出版日期:2020-01-01

作者单位    点击查看

侯春萍:天津大学电气自动化与信息工程学院, 天津 300072
蒋天丽:天津大学电气自动化与信息工程学院, 天津 300072
郎玥:天津大学电气自动化与信息工程学院, 天津 300072
杨阳:天津大学电气自动化与信息工程学院, 天津 300072

联系人作者:郎玥(langyue@tju.edu.cn)

备注:国家自然科学基金;

【1】Fairchild D P, Narayanan R M. Classification of human motions using empirical mode decomposition of human micro-Doppler signatures [J]. IET Radar, Sonar & Navigation. 2014, 8(5): 425-434.

【2】Zhang Y, Zhu Y P, Cheng Y Q, et al. Human target radar echo signal analysis based on micro-Doppler characteristic [J]. Signal Processing. 2009, 25(10): 1616-1623.
张翼, 朱玉鹏, 程永强, 等. 基于微多普勒特征的人体目标雷达回波信号分析 [J]. 信号处理. 2009, 25(10): 1616-1623.

【3】Huang Y W, Wan C L, Feng H. Multi-feature fusion human behavior recognition algorithm based on convolutional neural network and long short term memory neural network [J]. Laser & Optoelectronics Progress. 2019, 56(7): 071505.
黄友文, 万超伦, 冯恒. 基于卷积神经网络与长短期记忆神经网络的多特征融合人体行为识别算法 [J]. 激光与光电子学进展. 2019, 56(7): 071505.

【4】Ikeda T, Ishiguro H, Miyashita T, et al. Pedestrian identification by associating wearable and environmental sensors based on phase dependent correlation of human walking [J]. Journal of Ambient Intelligence and Humanized Computing. 2014, 5(5): 645-654.

【5】Xu H Y, Kong J, Jiang M, et al. Action recognition based on histogram of spatio-temporal oriented principal components [J]. Laser & Optoelectronics Progress. 2018, 55(6): 061009.
徐海洋, 孔军, 蒋敏, 等. 基于时空方向主成分直方图的人体行为识别 [J]. 激光与光电子学进展. 2018, 55(6): 061009.

【6】van Dorp P, Groen F C A. Feature-based human motion parameter estimation with radar [J]. IET Radar, Sonar & Navigation. 2008, 2(2): 135-145.

【7】Kim Y, Ling H. Human activity classification based on micro-Doppler signatures using a support vector machine [J]. IEEE Transactions on Geoscience and Remote Sensing. 2009, 47(5): 1328-1337.

【8】Ricci R, Balleri A. Recognition of humans based on radar micro-Doppler shape spectrum features [J]. IET Radar, Sonar & Navigation. 2015, 9(9): 1216-1223.

【9】Li J, Phung S L. Tivive F H C, et al. Automatic classification of human motions using Doppler radar . [C]∥The 2012 International Joint Conference on Neural Networks (IJCNN), June 10-15, 2012, Brisbane, QLD, Australia. New York: IEEE. 2012, 12906230.

【10】Javier R J, Kim Y. Application of linear predictive coding for human activity classification based on micro-Doppler signatures [J]. IEEE Geoscience and Remote Sensing Letters. 2014, 11(10): 1831-1834.

【11】Tahmoush D. Review of micro-Doppler signatures [J]. IET Radar, Sonar & Navigation. 2015, 9(9): 1140-1146.

【12】Kim Y, Moon T. Human detection and activity classification based on micro-Doppler signatures using deep convolutional neural networks [J]. IEEE Geoscience and Remote Sensing Letters. 2016, 13(1): 8-12.

【13】Cao P B, Xia W J, Ye M, et al. Radar-ID: human identification based on radar micro-Doppler signatures using deep convolutional neural networks [J]. IET Radar, Sonar & Navigation. 2018, 12(7): 729-734.

【14】Chen Z X, Li G, Fioranelli F, et al. Personnel recognition and gait classification based on multistatic micro-Doppler signatures using deep convolutional neural networks [J]. IEEE Geoscience and Remote Sensing Letters. 2018, 15(5): 669-673.

【15】Seyfioglu M S, Ozbayoglu A M, Gurbuz S Z. Deep convolutional autoencoder for radar-based classification of similar aided and unaided human activities [J]. IEEE Transactions on Aerospace and Electronic Systems. 2018, 54(4): 1709-1723.

【16】Kim Y, Li Y. Human activity classification with transmission and reflection coefficients of on-body antennas through deep convolutional neural networks [J]. IEEE Transactions on Antennas and Propagation. 2017, 65(5): 2764-2768.

【17】Klarenbeek G. Harmanny R I A, Cifola L. Multi-target human gait classification using LSTM recurrent neural networks applied to micro-Doppler . [C]∥2017 European Radar Conference (EURAD), October 11-13, 2017, Nuremberg, Germany. New York: IEEE. 2017, 167-170.

【18】Lu Y X, Kumar A, Zhai S F, et al. Fully-adaptive feature sharing in multi-task networks with applications in person attribute classification . [C]∥2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), July 21-26, 2017, Honolulu, HI, USA. New York: IEEE. 2017, 1131-1140.

【19】Long M S, Cao Z J, Wang J M, et al. Learning multiple tasks with multilinear relationship networks . [C]∥Advances in Neural Information Processing Systems 30 (NIPS 2017), December 4-9, 2017, Long Beach, CA, USA. Canada: NIPS. 2017, 1594-1603.

【20】Hashimoto K, Xiong C M, Tsuruoka Y, et al. -07-24)[2019-06-09] . https:∥arxiv. 2017, org/abs/1611: 01587.

【21】Kendall A, Gal Y, Cipolla R. Multi-task learning using uncertainty to weigh losses for scene geometry and semantics . [C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, June 18-22, 2018, Salt Lake City, Utah. New York: IEEE. 2018, 7482-7491.

【22】Liu Y P, Wang C, Xia H Y. Application progress of time-frequency analysis for lidar [J]. Laser & Optoelectronics Progress. 2018, 55(12): 120005.
刘燕平, 王冲, 夏海云. 时频分析在激光雷达中的应用进展 [J]. 激光与光电子学进展. 2018, 55(12): 120005.

【23】Hu J, Shen L, Albanie S, et al. Squeeze-and-excitation networks [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2019, 1.

【24】Zhou G B, Wu J X, Zhang C L, et al. Minimal gated unit for recurrent neural networks [J]. International Journal of Automation and Computing. 2016, 13(3): 226-234.

【25】Wen Y D, Zhang K P, Li Z F, et al. A discriminative feature learning approach for deep face recognition [M]. ∥Leibe B, Matas J, Sebe N, et al. Computer vision-ECCV 2016. Lecture notes in computer science. Cham: Springer. 2016, 9911: 499-515.

引用该论文

Hou Chunping,Jiang Tianli,Lang Yue,Yang Yang. Human Activity and IdentityMulti-Task Recognition Based on Convolutional Neural Network Using Doppler Radar[J]. Laser & Optoelectronics Progress, 2020, 57(2): 021009

侯春萍,蒋天丽,郎玥,杨阳. 基于卷积神经网络的雷达人体动作与身份多任务识别[J]. 激光与光电子学进展, 2020, 57(2): 021009

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF

欧洲女人性开放视频