Robust Single Accelerometer-Based Activity Recognition Using Modified Recurrence Plot

Using a single 3-axis accelerometer for human activity recognition is challenging but attractive in daily healthcare and monitoring with wearable sensors and devices. In this paper, an effective and efficient framework is proposed to address the recognition problem without any heavy preprocessing. We encode 3-axis signals as 3-channel images using a modified recurrence plot (RP) and train a tiny residual neural network to do image classification. The modified RP is first proposed in our paper to overcome its tendency confusion problem, which has improved our system performance significantly. We evaluate our algorithm on a new database and a public dataset. Results show that our recognition framework achieves highly competitive accuracies and good efficiencies with other state-of-the-art methods on both datasets. Moreover, our method shows stronger robustness to noise and low decimation rate through comparison experiments. Finally, we provide detailed discussion and analysis of our approach from two perspectives: the pattern analysis of encoding algorithm and the interpretation of classification model.

[1]  D. Ruelle,et al.  Recurrence Plots of Dynamical Systems , 1987 .

[2]  Ferran Sanz,et al.  Identifying temporal patterns in patient disease trajectories using dynamic time warping: A population-based study , 2018, Scientific Reports.

[3]  Xiaoli Li,et al.  Deep Convolutional Neural Networks on Multichannel Time Series for Human Activity Recognition , 2015, IJCAI.

[4]  Norbert Marwan,et al.  A historical review of recurrence plots , 2008, 1709.09971.

[5]  Qingquan Li,et al.  Robust Gait Recognition by Integrating Inertial and RGBD Sensors , 2016, IEEE Transactions on Cybernetics.

[6]  Tim Oates,et al.  Encoding Time Series as Images for Visual Inspection and Classification Using Tiled Convolutional Neural Networks , 2014 .

[7]  Jerker Westin,et al.  A Treatment-Response Index From Wearable Sensors for Quantifying Parkinson's Disease Motor States , 2018, IEEE Journal of Biomedical and Health Informatics.

[8]  Fulvio Mastrogiovanni,et al.  A public domain dataset for ADL recognition using wrist-placed accelerometers , 2014, The 23rd IEEE International Symposium on Robot and Human Interactive Communication.

[9]  Eamonn J. Keogh,et al.  The great time series classification bake off: a review and experimental evaluation of recent algorithmic advances , 2016, Data Mining and Knowledge Discovery.

[10]  Eamonn J. Keogh,et al.  Extracting Optimal Performance from Dynamic Time Warping , 2016, KDD.

[11]  Paolo Dario,et al.  Toward an Unsupervised Approach for Daily Gesture Recognition in Assisted Living Applications , 2017, IEEE Sensors Journal.

[12]  Eamonn J. Keogh,et al.  Searching and Mining Trillions of Time Series Subsequences under Dynamic Time Warping , 2012, KDD.

[13]  Eamonn J. Keogh,et al.  Speeding up similarity search under dynamic time warping by pruning unpromising alignments , 2018, Data Mining and Knowledge Discovery.

[14]  Shahrokh Valaee,et al.  A Novel Accelerometer-based Gesture Recognition System by , 2010 .

[15]  Houshang Darabi,et al.  LSTM Fully Convolutional Networks for Time Series Classification , 2017, IEEE Access.

[16]  Ilkka Korhonen,et al.  An Activity Recognition Framework Deploying the Random Forest Classifier and A Single Optical Heart Rate Monitoring and Triaxial Accelerometer Wrist-Band † , 2018, Sensors.

[17]  Igor Bisio,et al.  Enabling IoT for In-Home Rehabilitation: Accelerometer Signals Classification Methods for Activity and Movement Recognition , 2017, IEEE Internet of Things Journal.

[18]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[19]  Vinícius M. A. de Souza,et al.  Time Series Classification Using Compression Distance of Recurrence Plots , 2013, 2013 IEEE 13th International Conference on Data Mining.

[20]  Sajal K. Das,et al.  HARKE: Human Activity Recognition from Kinetic Energy Harvesting Data in Wearable Devices , 2018, IEEE Transactions on Mobile Computing.

[21]  Macarena Espinilla,et al.  Sensor-Based Datasets for Human Activity Recognition – A Systematic Review of Literature , 2018, IEEE Access.

[22]  Tim Oates,et al.  Time series classification from scratch with deep neural networks: A strong baseline , 2016, 2017 International Joint Conference on Neural Networks (IJCNN).

[23]  Giancarlo Fortino,et al.  Recognition of human fall events based on single tri-axial gyroscope , 2018, 2018 IEEE 15th International Conference on Networking, Sensing and Control (ICNSC).

[24]  Jürgen Kurths,et al.  Recurrence plots for the analysis of complex systems , 2009 .

[25]  Reza Malekian,et al.  Physical Activity Recognition From Smartphone Accelerometer Data for User Context Awareness Sensing , 2017, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[26]  William Robson Schwartz,et al.  Human Activity Recognition Based on Wearable Sensor Data: A Standardization of the State-of-the-Art , 2018, ArXiv.

[27]  Vincenzo Della Mea,et al.  A feasibility study on smartphone accelerometer-based recognition of household activities and influence of smartphone position , 2017, Informatics for health & social care.

[28]  Bolei Zhou,et al.  Learning Deep Features for Discriminative Localization , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[29]  Nima Hatami,et al.  Classification of time-series images using deep convolutional neural networks , 2017, International Conference on Machine Vision.

[30]  Xuefeng Chen,et al.  Dislocated Time Series Convolutional Neural Architecture: An Intelligent Fault Diagnosis Approach for Electric Machine , 2017, IEEE Transactions on Industrial Informatics.