Please wait a minute...
Journal of ZheJiang University (Engineering Science)  2019, Vol. 53 Issue (6): 1130-1138    DOI: 10.3785/j.issn.1008-973X.2019.06.012
Mechanical and Energy Engineering     
Design of elderly fall detection system using CNN
Yan LV1,2(),Meng ZHANG1,Wu-hao JIANG1,Yi-hua NI1,*(),Xiao-hong QIAN3
1. School of Engineering, Zhejiang A&F University, Lin’an 311300, China
2. College of Mechanical Engineering, Zhejiang University, Hangzhou 310027, China
3. Enjoyor Research Institute, Enjoyor Co. Ltd, Hangzhou 310030, China
Download: HTML     PDF(1393KB) HTML
Export: BibTeX | EndNote (RIS)      

Abstract  

A fall detection model based on convolutional neural network (CNN) was constructed, in order to use portable device to accurately detect the fall of the elderly and to avoid the incompleteness caused by the artificial designed features in traditional algorithms. The three-axis sensor built in the smart phone was used as the data acquisition source; the collected human body posture information was filtered, standardized, and sampled, etc. and then inputted into the designed model. Multi-layer CNN was used to train and optimize the key parameters of the model, combined with gradient descent and adaptive moment estimation optimization methods; the learned deep features were used for samples, classification. The experimental results show that the accuracy of the designed model for fall detection is significantly higher than that of the general machine learning algorithm model. In addition, the evaluation indicators have maintained a high level of stability in the detection of falls and non-falls.



Key wordsfall detection      smart phone sensor      convolutional neural network (CNN)      deep learning     
Received: 04 May 2018      Published: 22 May 2019
CLC:  TP 391  
Corresponding Authors: Yi-hua NI     E-mail: lvyan@zju.edu.cn;nyh@zafu.edu.cn
Cite this article:

Yan LV,Meng ZHANG,Wu-hao JIANG,Yi-hua NI,Xiao-hong QIAN. Design of elderly fall detection system using CNN. Journal of ZheJiang University (Engineering Science), 2019, 53(6): 1130-1138.

URL:

http://www.zjujournals.com/eng/10.3785/j.issn.1008-973X.2019.06.012     OR     http://www.zjujournals.com/eng/Y2019/V53/I6/1130


采用卷积神经网络的老年人跌倒检测系统设计

为了利用便携式设备准确检测老年人的跌倒状况,针对传统算法中人为设计特征造成的不完备性,构建一种基于卷积神经网络(CNN)的老年人跌倒检测模型. 以智能手机内置的三轴传感器作为数据获取源,将采集的人体姿态信息进行滤波、标准化、采样等操作后,输入到所设计的模型中;采用梯度下降和适应性动量优化方法进行多层卷积神经网络训练和优化,获得模型关键参数训练并优化模型关键参数;利用学习到的深层次特征进行样本分类. 实验结果表明:所设计的模型对于跌倒检测的准确率明显高于一般的机器学习算法模型,并且在对跌倒和非跌倒的区分检测中,精确率和召回率都保持了较高的稳定水平.


关键词: 跌倒检测,  手机传感器,  卷积神经网络(CNN),  深度学习 
Fig.1 Elderly fall detection process using convolutional neural network (CNN)
Fig.2 CNN model structure for elderly fall detection
Fig.3 Acquisition process of mobile phone sensor data (Dataset1)
Fig.4 Human 3D motion data acquisition equipments
Fig.5 Training sample data format conversion for fall detection
数据集 跌倒样本 非跌倒样本
采样前 采样后 采样前 采样后
Dataset1 328 656 2 072 656
Dataset2 845 1 690 4 155 1 690
Tab.1 Sample size distribution before and after sampling
Fig.6 Structure design of fall detection model using CNN
Fig.7 Fall detection model training data set training results (Dataset1)
Fig.8 Fall detection model training data set training results (Dataset2)
Fig.9 Detection accuracy of different classification models on fall datasets Dataset1 and Dataset2
%
分类模型 P R F1 A
Dataset1 Dataset2 Dataset1 Dataset2 Dataset1 Dataset2 Dataset1 Dataset2
SVM 93.4 76.8 82.8 92.1 87.8 83.7 88.1 82.2
DT 90.0 80.9 87.9 68.1 88.9 73.9 88.8 76.1
RF 91.3 89.0 88.6 88.5 89.9 88.8 89.8 88.8
GBDT 90.8 86.1 87.0 86.6 88.8 86.4 88.7 86.4
AdaBoost 87.6 79.9 82.7 82.5 85.0 81.2 85.0 80.9
Voting 91.4 84.7 86.7 88.0 89.0 86.3 88.9 86.1
CNN 99.5 100 98.5 99.5 98.9 99.7 99.2 99.8
Tab.2 Comparison of each models detection results using fall datasets Dataset1 and Dataset2
Fig.10 Comparison of accuracy and recall of each model using fall dataset Dataset1
Fig.11 Comparison of accuracy and recall of each model using fall dataset Dataset2
[1]   师昉, 李福亮, 孙思佳, 等 中国老年跌倒研究的现状与对策[J]. 中国康复, 2018, 33 (3): 246- 248
SHI Fang, LI Fu-liang, SUN Si-jia, et al The present situation and countermeasures of the study of senile falls in China[J]. Chinese Journal of Rehabilitation, 2018, 33 (3): 246- 248
[2]   陈瑜, 姜小鹰, 杨美丽, 等 养老机构老年人跌倒效能状况及其影响因素分析[J]. 中国实用护理, 2016, 32 (6): 412- 415
CHEN Yu, JIANG Xiao-ying, YANG Mei-li, et al Investigation on the fall efficacy of elderly people in nursing homes and its influencing factors[J]. Chinese Journal of Practical Nursing, 2016, 32 (6): 412- 415
doi: 10.3760/cma.j.issn.1672-7088.2016.06.004
[3]   KHAN S S, HOEY J Review of fall detection techniques: a data availability perspective[J]. Medical Engineering and Physics, 2017, 1 (39): 12- 22
[4]   GHEN K H, HSU Y W, YANG J J, et al Enhanced characterization of an accelerometer-based fall detection algorithm using a repository[J]. Instrumentation Science and Technology, 2016, 11 (6): 382- 391
[5]   MADARSHAHIAN R, CAICEDO J M, ZAMBRANA D A. Evaluation of a time reversal method with dynamic time warping matching function for human fall detection using structural vibrations [C] // Model Validation and Uncertainty Quantification, Volume 3, Conference Proceedings of the Society for Experimental Mechanics Series. Florida: Springer, 2014: 171-176.
[6]   CHEN K H, YANG J J, JAW F S Accelerometer-based fall detection using feature extraction and support vector machine algorithms[J]. Instrumentation Science and Technology, 2016, 44 (2): 333- 342
[7]   LIM D, PARK C, KIM N Fall-detection algorithm using 3-axis acceleration: combination with simple threshold and hidden markov model[J]. Journal of Applied Mathematics, 2014, 9 (17): 342- 351
[8]   AZIZ O, MUSNGI M, PARK E J, et al. A comparison of accuracy of fall detection algorithms (threshold-based vs. machine learning) using waist-mounted tri-axial accelerometer signals from a comprehensive set of falls and non-fall trials[J]. Medical and Biological Engineering and Computing, 2017, 55 (1): 45- 55
doi: 10.1007/s11517-016-1504-y
[9]   OZDEMIR A, BARSHAN B Detecting falls with wearable sensors using machine learning techniques[J]. Sensors, 2014, 14 (6): 10691- 10708
doi: 10.3390/s140610691
[10]   MONGIA M, KUMAR K, ERRAQABI A, et al. On random weights for texture generation in one layer CNNS [C] // ICASSP IEEE International Conference on Acoustics. New Orleans: IEEE, 2017: 2207-2211.
[11]   XIA Y, NAREN W L, WANG K Q, et al Detecting atrial fibrillation by deep convolutional neural networks[J]. Computers in Biology and Medicine, 2018, 93 (2): 84- 92
[12]   ABADI M, BARHAM P, CHEN J M, et al. TensorFlow: a system for large-scale machine learning [C] // Proceedings of the12th USENIX Symposium on Operating Systems Design and Implementation. Savannah: USENIX, 2016: 265-283.
[1] Jia-hui XU,Jing-chang WANG,Ling CHEN,Yong WU. Surface water quality prediction model based on graph neural network[J]. Journal of ZheJiang University (Engineering Science), 2021, 55(4): 601-607.
[2] Hong-li WANG,Bin GUO,Si-cong LIU,Jia-qi LIU,Yun-gang WU,Zhi-wen YU. End context-adaptative deep sensing model with edge-end collaboration[J]. Journal of ZheJiang University (Engineering Science), 2021, 55(4): 626-638.
[3] Teng ZHANG,Xin-long JIANG,Yi-qiang CHEN,Qian CHEN,Tao-mian MI,Piu CHAN. Wrist attitude-based Parkinson's disease ON/OFF state assessment after medication[J]. Journal of ZheJiang University (Engineering Science), 2021, 55(4): 639-647.
[4] Li-feng XU,Hai-fan HUANG,Wei-long DING,Yu-lei FAN. Detection of small fruit target based on improved DenseNet[J]. Journal of ZheJiang University (Engineering Science), 2021, 55(2): 377-385.
[5] Hao-can XU,Ji-tuo LI,Guo-dong LU. Reconstruction of three-dimensional human bodies from single image by LeNet-5[J]. Journal of ZheJiang University (Engineering Science), 2021, 55(1): 153-161.
[6] Yi-peng HUANG,Ji-su HU,Xu-sheng QIAN,Zhi-yong ZHOU,Wen-lu ZHAO,Qi MA,Jun-kang SHEN,Ya-kang DAI. SE-Mask-RCNN: segmentation method for prostate cancer on multi-parametric MRI[J]. Journal of ZheJiang University (Engineering Science), 2021, 55(1): 203-212.
[7] Pu ZHENG,Hong-yang BAI,Wei LI,Hong-wei GUO. Small target detection algorithm in complex background[J]. Journal of ZheJiang University (Engineering Science), 2020, 54(9): 1777-1784.
[8] Qiao-hong CHEN,YI CHEN,Wen-shu Li,Yu-bo JIA. Clothing image classification based on multi-scale SE-Xception[J]. Journal of ZheJiang University (Engineering Science), 2020, 54(9): 1727-1735.
[9] Deng-wen ZHOU,Jin-yue TIAN,Lu-yao MA,Xiu-xiu SUN. Lightweight image semantic segmentation based on multi-level feature cascaded network[J]. Journal of ZheJiang University (Engineering Science), 2020, 54(8): 1516-1524.
[10] Tao MING,Dan WANG,Ji-chang GUO,Qiang LI. Breast cancer histopathological image classification using multi-scale channel squeeze-and-excitation model[J]. Journal of ZheJiang University (Engineering Science), 2020, 54(7): 1289-1297.
[11] Xu YAN,Xiao-liang FAN,Chuan-pan ZHENG,Yu ZANG,Cheng WANG,Ming CHENG,Long-biao CHEN. Urban traffic flow prediction algorithm based on graph convolutional neural networks[J]. Journal of ZheJiang University (Engineering Science), 2020, 54(6): 1147-1155.
[12] Ping YANG,Dan WANG,Zi-jian KAGN,Tong LI,Li-hua FU,Yue-ren YU. Prediction model of paroxysmal atrial fibrillation based on pattern recognition and ensemble CNN-LSTM[J]. Journal of ZheJiang University (Engineering Science), 2020, 54(5): 1039-1048.
[13] Zhou-fei WANG,Wei-na YUAN. Channel estimation and detection method for multicarrier system based on deep learning[J]. Journal of ZheJiang University (Engineering Science), 2020, 54(4): 732-738.
[14] Bing YANG,Wen-bo MO,Jin-liang YAO. 3D palmprint recognition by using local features and deep learning[J]. Journal of ZheJiang University (Engineering Science), 2020, 54(3): 540-545.
[15] Yan-jia HONG,Tie-bao MENG,Hao-jiang LI,Li-zhi LIU,Li LI,Shuo-yu XU,Sheng-wen GUO. Deep segmentation method of tumor boundaries from MR images of patients with nasopharyngeal carcinoma using multi-modality and multi-dimension fusion[J]. Journal of ZheJiang University (Engineering Science), 2020, 54(3): 566-573.