基于SE-CNN的人体摔倒检测方法

doi:10.19678/j.issn.1000-3428.0061833

摘要/Abstract

摘要： 人口老龄化是当今社会发展不可忽视的问题，目前有很大一部分老年人在无人照顾的境况下独自生活，摔倒后无法及时得到救助成为威胁老人生命安全的重要原因之一。现有的人体摔倒检测方法存在适应性差、高入侵性、易误判、成本昂贵等问题，且无法快速、实时检测老人摔倒。提出一种基于机器学习和无线传感器网络的摔倒检测方法，使用多个物联网传感节点组建无线传感器网络采集RSS数据，对采集到的RSS数据进行预处理后，通过XGBoost模型对时域特征分量和小波域特征分量进行处理，并以排列组合方式得到具有强鲁棒性的联合特征分量。利用深度学习网络获得数据潜在规律的特点构建人体摔倒识别模型，采用卷积神经网络作为主干网络，并在相邻网络层之间引入通道注意力模块，通过构建SE-CNN模型实现人体摔倒检测。实验结果表明，联合特征的加入能够提高RSS数据的可区分性，且SE-CNN模型的识别准确率高于CNN模型，可以实现高准确率的人体摔倒检测。

关键词: 摔倒检测, 无线传感, 小波变换, 联合特征, SE-CNN模型

Abstract: Population aging is a problem that cannot be ignored in today's social development.When a large part of the elderly live alone without someone taking care of them, the failure to receive timely assistance after a fall has become one reason that threatens the safety of the elderly.Existinghuman fall detection methods have problems such as poor adaptability, high invasiveness, easy misjudgment, and high cost.Moreover, they cannot quickly and accurately detect falls in the elderly.A fall detection method based on machine learning and wireless sensor networks is proposed so family members can discover the elderly living alone if they fall.A wireless sensor network is formed by multiple internet of things sensor nodes to collect Received Signal Strength(RSS) data;secondly, after preprocessing the collected RSS data, the XGBoost model permutates and combines the time-domain feature components and wavelet domain feature components to obtain strong Robust joint feature components.This study uses a deep learning network to learn the characteristics of the potential laws of the data to build an optimal human fall recognition model, andthe Convolutional Neural Network (CNN) is used as the backbone network, the Squeeze and Excitation (SE) module is introduced between the adjacent network layers to realize the human fall detection by constructing the SE-CNN model.Experiments show that the addition of joint features improves the distinguishability of the RSS data, and the recognition accuracy of the SE-CNN model is higher than that of the CNN model, which detects human falls with high accuracy.

Key words: fall detection, wireless sensing, wavelet transform, joint feature, SE-CNN model

中图分类号:

TP391

杨志勇, 王俊杰, 金磊. 基于SE-CNN的人体摔倒检测方法[J]. 计算机工程, 2022, 48(6): 270-277.

YANG Zhiyong, WANG Junjie, JIN Lei. Human Fall Detection Method Based on SE-CNN[J]. Computer Engineering, 2022, 48(6): 270-277.

http://www.ecice06.com/CN/Y2022/V48/I6/270

图/表 17

20220625181149

20220625181153

20220625181156

20220625181200

20220625181204

20220625181207

20220625181212

20220625181217

20220625181220

20220625181224

20220625181227

20220625181231

20220625181235

20220625181239

20220625181243

20220625181247

20220625181252

参考文献

[1] SOLBACH M D, TSOTSOS J K.Vision-based fallen person detection for the elderly[C]//Proceedings of IEEE International Conference on Computer Vision.Washington D.C., USA:IEEE Press, 2017:1433-1442.
[2] SANTOS G, ENDO P, MONTEIRO K, et al.Accelerometer-based human fall detection using convolutional neural networks[J].Sensors, 2019, 19(7):1644-1656.
[3] HSIEH C Y, LIU K C, HUANG C N, et al.Novel hierarchical fall detection algorithm using a multiphase fall model[J].Sensors, 2017, 17(2):307-312.
[4] WU Y F, SU Y W, FENG R J, et al.Wearable-sensor-based pre-impact fall detection system with a hierarchical classifier[J].Measurement, 2019, 140:283-29.
[5] WANG F T, CHAN H L, HSU M H, et al.Threshold-based fall detection using a hybrid of tri-axial accelerometer and gyroscope[J].Physiological Measurement, 2018, 39(10):105002-105011.
[6] VAN SCHOOTEN K S, PIJNAPPELS M, RISPENS S M, et al.Daily-life gait quality as predictor of falls in older people:a 1-year prospective cohort study[J].PLoS One, 2016, 11(7):23-32.
[7] MAO A H, MA X D, HE Y N, et al.Highly portable, sensor-based system for human fall monitoring[J].Sensors, 2017, 17(9):2096-2103.
[8] OZCAN K, MAHABALAGIRI A K, CASARES M, et al.Automatic fall detection and activity classification by a wearable embedded smart camera[J].IEEE Journal on Emerging and Selected Topics in Circuits and Systems, 2013, 3(2):125-136.
[9] OZCAN K, VELIPASALAR S, VARSHNEY P K.Autonomous fall detection with wearable cameras by using relative entropy distance measure[J].IEEE Transactions on Human-Machine Systems, 2017, 47(1):31-39.
[10] OZCAN K, VELIPASALAR S.Wearable camera-and accelerometer-based fall detection on portable devices[J].IEEE Embedded Systems Letters, 2016, 8(1):6-9.
[11] SINGH A, REHMAN S U, YONGCHAREON S, et al.Sensor technologies for fall detection systems:a review[J].IEEE Sensors Journal, 2020, 20(13):6889-6919.
[12] POPESCU M, HOTRABHAVANANDA B, MOORE M, et al.VAMPIR-an automatic fall detection system using a vertical PIR sensor array[C]//Proceedings of the 6th International Conference on Pervasive Computing Technologies for Healthcare and Workshops.Washington D.C., USA:IEEE Press, 2012:163-166.
[13] ER P V, TAN K K.Non-intrusive fall detection monitoring for the elderly based on fuzzy logic[J].Measurement, 2018, 124:91-102.
[14] ALWAN M, RAJENDRAN P J, KELL S, et al.A smart and passive floor-vibration based fall detector for elderly[C]//Proceedings of the 2nd International Conference on Information & Communication Technologies.Washington D.C., USA:IEEE Press, 2006:1003-1007.
[15] GUAN Q J, LI C Y, GUO X M, et al.Infrared signal based elderly fall detection for in-home monitoring[C]//Proceedings of the 9th International Conference on Intelligent Human-Machine Systems and Cybernetics.Washington D.C., USA:IEEE Press, 2017:373-376.
[16] ZIGEL Y, LITVAK D, GANNOT I.A method for automatic fall detection of elderly people using floor vibrations and sound-proof of concept on human mimicking doll Falls[J].IEEE Transactions on Bio-Medical Engineering, 2009, 56(12):2858-2867.
[17] HARROU F, ZERROUKI N, SUN Y, et al.Vision-based fall detection system for improving safety of elderly people[J].IEEE Instrumentation & Measurement Magazine, 2017, 20(6):49-55.
[18] ROUGIER C, MEUNIER J, ST-ARNAUD A, et al.Fall detection from human shape and motion history using video surveillance[C]//Proceedings of the 21st International Conference on Advanced Information Networking and Applications Workshops.Washington D.C., USA:IEEE Press, 2007:875-880.
[19] GUNALE K G, MUKHERJI P.Fall detection using k-nearest neighbor classification for patient monitoring[C]//Proceedings of International Conference on Information Processing.Washington D.C., USA:IEEE Press, 2015:520-524.
[20] SHOTTON J, GIRSHICK R, FITZGIBBON A, et al.Efficient human pose estimation from single depth images[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(12):2821-2840.
[21] SASE P S, BHANDARI S H.Human fall detection using depth videos[C]//Proceedings of the 5th International Conference on Signal Processing and Integrated Networks.Washington D.C., USA:IEEE Press, 2018:546-549.
[22] BALDEWIJNS G, DEBARD G, MERTES G, et al.Improving the accuracy of existing camera based fall detection algorithms through late fusion[C]//Proceedings of the 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.Washington D.C., USA:IEEE Press, 2017:2667-2671.
[23] SASE P S, BHANDARI S H.Human fall detection using depth videos[C]//Proceedings of the 5th International Conference on Signal Processing and Integrated Networks.Washington D.C., USA:IEEE Press, 2018:546-549.
[24] SU B Y, HO K C, RANTZ M J, et al.Doppler radar fall activity detection using the wavelet transform[J].IEEE Transactions on Bio-Medical Engineering, 2015, 62(3):865-875.
[25] WANG B, GUO L, ZHANG H, et al.A millimetre-wave radar-based fall detection method using line kernel convolutional neural network[J].IEEE Sensors Journal, 2020, 20(22):13364-13370.
[26] WANG H, ZHANG D Q, WANG Y S, et al.RT-fall:a real-time and contactless fall detection system with commodity WiFi devices[J].IEEE Transactions on Mobile Computing, 2017, 16(2):511-526.
[27] KEATON M, NORDSTROM A, SHERWOOD M, et al.WiFi-based in-home fall-detection utility:application of WiFi channel state information as a fall detection service[C]//Proceedings of IEEE International Conference on Engineering, Technology and Innovation.Washington D.C., USA:IEEE Press, 2020:1-6.
[28] HU Y Q, ZHANG F, WU C S, et al.A WiFi-based passive fall detection system[C]//Proceedings of ICASSP IEEE International Conference on Acoustics, Speech and Signal Processing.Washington D.C., USA:IEEE Press, 2020:1723-1727.
[29] SADREAZAMI H, BOLIC M, RAJAN S.Fall detection using standoff radar-based sensing and deep convolutional neural network[J].IEEE Transactions on Circuits and Systems II:Express Briefs, 2020, 67(1):197-201.
[30] DING J Y, WANG Y.A WiFi-based smart home fall detection system using recurrent neural network[J].IEEE Transactions on Consumer Electronics, 2020, 66(4):308-317.
[31] 霍宏伟, 张宏科.基于室内无线传感器网络射频信号的老年人跌倒检测研究[J].电子学报, 2011, 39(1):195-200. HUO H W, ZHANG H K.Fall detection using radio signals of home wireless sensor networks[J].Acta Electronica Sinica, 2011, 39(1):195-200.(in Chinese)
[32] LUO K, JIN X F.Fall recognition approach based on human skeleton information[C]//Proceedings of the 6th International Conference on Instrumentation & Measurement, Computer, Communication and Control.Washington D.C., USA:IEEE Press, 2016:707-711.
[33] SHIBA K, KABURAGI T, KURIHARA Y.Fall detection utilizing frequency distribution trajectory by microwave Doppler sensor[J].IEEE Sensors Journal, 2017, 17(22):7561-7568.
[34] WANG Y X, WU K S, NI L M.WiFall:device-free fall detection by wireless networks[J].IEEE Transactions on Mobile Computing, 2017, 16(2):581-594.

选择文件类型/文献管理软件名称

选择包含的内容