International Journal of Multimedia Computing, 2022, 3(2); doi: 10.38007/IJMC.2022.030205.
Xianmin Ma
Department of Information Engineering, Heilongjiang International University, Heilongjiang, China
In order to improve the accuracy and automation of daily activity recognition and reduce manual intervention, a deep neural network based on wearable sensor signals was proposed for human activity recognition. This paper aims to identify and predict human fall behavior based on wearable inertial sensors. Selecting different device locations can have a significant impact on the efficiency of data collection. In this study, the data obtained by using sensors connected to the lower extremities should be much more effective than data obtained through the arms or other parts of the body when collecting exercise data related to the running behavior of the elderly. In addition to considering the validity of the data, we also need to consider how the data is related to the real situation and how comfortable the user can wear it in real life. Experimental data show that in order to obtain data sets that are easy to calculate, accurate and effective, two aspects need to be considered: the location of data collection equipment and the frequency of data collection. The experimental results show that the recognition accuracy is up to 93.7% after using the training decision tree algorithm with 562 features selected specifically for activity recognition. Then, the extracted features are trained by decision tree algorithm, and the recognition accuracy is only 82.8%. Finally, lSTM-RNN was used to directly train the original sensor data, and the accuracy of behavioral activity recognition in the elderly could reach 92.28%. In the aspect of behavior recognition, the traditional work mainly focuses on the identification of simple behaviors performed by a single person successively. But in real life, human behavior is complex, and applications often require high real-time recognition results.
Wearable, Inertial Sensors, Simple Behavior Recognition
Xianmin Ma. 5G Virtual Reality Internet of Things - Research on Human Fall Behavior Recognition and Prediction Based on Wearable Inertial Sensors. International Journal of Multimedia Computing (2022), Vol. 3, Issue 2: 29-41. https://doi.org/10.38007/IJMC.2022.030205.
[1] Shen Y, Phan N H, Xiao X, et al. Dynamic socialized Gaussian process models for human behavior prediction in a health social network. Knowledge and Information Systems. (2016) 49(2): 455-479. https://doi.org/10.1007/s10115-015-0910-z
[2] Xu M, Guo H, Hou C, et al. Corrigendum: Olfactory perception and behavioral effects of sex pheromone gland components in Helicoverpa armigera and Helicoverpa assulta. Scientific Reports. (2016) 6(1): 22998. https://doi.org/10.1038/srep22998
[3] Xu M, Guo H, Hou C, et al. Corrigendum: Olfactory perception and behavioral effects of sex pheromone gland components in Helicoverpa armigera and Helicoverpa assulta. Scientific Reports. (2016) 6(1): 22998. https://doi.org/10.1038/srep22998
[4] Dong X, Zhang L, Milholland B, et al. Accurate identification of single-nucleotide variants in whole-genome-amplified single cells. Nature Methods. (2017) 14(5): 491-493. https://doi.org/10.1038/nmeth.4227
[5] Li G, Liu T, Yi J, et al. The Lower Limbs Kinematics Analysis by Wearable Sensor Shoes. IEEE Sensors Journal. (2016) 16(8): 2627-2638. https://doi.org/10.1109/JSEN.2016.2515101
[6] Hayashi T, Tamura Y, Hasegawa T, et al. Record-Low Spatial Mode Dispersion and Ultra-Low Loss Coupled Multi-Core Fiber for Ultra-Long-Haul Transmission. Journal of Lightwave Technology. (2017) 35(3): 450-457. https://doi.org/10.1109/JLT.2016.2614000
[7] Souri Yaser, Noury Erfan, AdeliMosabbeb Ehsan. Deep Relative Attributes. IEEE transactions on multimedia. (2016) 18(9): 1832-1842. https://doi.org/10.1109/TMM.2016.2582379
[8] Bastug E, Bennis M, Medard M, et al. Toward Interconnected Virtual Reality: Opportunities, Challenges, and Enablers. IEEE Communications Magazine. (2017) 55(6): 110-117. https://doi.org/10.1109/MCOM.2017.1601089
[9] Elbamby M S, Perfecto C, Bennis M, et al. Towards Low-Latency and Ultra-Reliable Virtual Reality. IEEE Network. (2018) 32(2): 78-84. https://doi.org/10.1109/MNET.2018.1700268
[10] Boy J, Eveillard L, Detienne F, et al. Suggested Interactivity: Seeking Perceived Affordances for Information Visualization. IEEE transactions on visualization & computer graphics. (2016) 22(1): 639. https://doi.org/10.1109/TVCG.2015.2467201
[11] Berg L P, Vance J M. Industry use of virtual reality in product design and manufacturing: a survey. Virtual Reality. (2017) 21(1): 1-17. https://doi.org/10.1007/s10055-016-0293-9
[12] Chen X, Jiao L, Li W, et al. Efficient Multi-User Computation Offloading for Mobile-Edge Cloud Computing. IEEE/ACM Transactions on Networking. (2016) 24(5): 2795-2808. https://doi.org/10.1109/TNET.2015.2487344
[13] Batchuluun G, Kim J H, Hong H G, et al. Fuzzy system based human behavior recognition by combining behavior prediction and recognition. Expert Systems with Applications. (2017) 81(SEP.): 108-133. https://doi.org/10.1016/j.eswa.2017.03.052
[14] Guo B, Chen Y J, Lane N, et al. Behavior Recognition Based on Wi-Fi CSI: Part 1. IEEE Communications Magazine. (2017) 55(10): 90-90. https://doi.org/10.1109/MCOM.2017.8067691
[15] Olszewski K, Lim J J, Saito S, et al. High-fidelity facial and speech animation for VR HMDs. Acm Transactions on Graphics. (2016) 35(6cd): 221. https://doi.org/10.1145/2980179.2980252
[16] Browne G J, Cheung C M K, Heinzl A, et al. Human Information Behavior. Wirtschaftsinformatik. (2017) 59(1): 1-2. https://doi.org/10.1007/s12599-016-0458-9
[17] Yang L, Liu J, Zhang Y. An Intelligent Security Defensive Model of SCADA Based on Multi-Agent in Oil and Gas Fields. International Journal of Pattern Recognition and Artificial Intelligence. (2020) 34(01): 269-282. https://doi.org/10.1142/S021800142059003X
[18] Bhowmik T K, Parui S K, Roy U, et al. Bangla Handwritten Character Segmentation Using Structural Features: A Supervised and Bootstrapping Approach. Acm Transactions on Asian Language Information Processing. (2016) 15(4): 29.1-29.26. https://doi.org/10.1145/2890497
[19] Okuno T, Hardison D M. Perception-production link in L2 Japanese vowel duration: Training with technology. Language Learning & Technology. (2016) 20(2): 61-80.
[20] Wang X, Zhang W, Zhang Y, et al. Top-kspatial-keyword publish/subscribe over sliding window. Vldb Journal. (2017) 26(3): 301-326. https://doi.org/10.1007/s00778-016-0453-2
[21] Huimin Z, Meng S, Wu D, et al. A New Feature Extraction Method Based on EEMD and Multi-Scale Fuzzy Entropy for Motor Bearing. Entropy. (2016) 19(1): 14. https://doi.org/10.3390/e19010014
[22] Tao D, Jin L, Yuan Y, et al. Ensemble Manifold Rank Preserving for Acceleration-Based Human Activity Recognition. IEEE Transactions on Neural Networks and Learning Systems. (2016) 27(6): 1392-1404. https://doi.org/10.1109/TNNLS.2014.2357794
[23] Yasuhara M, Doi H, Wei C, et al. Biodiversity-ecosystem functioning relationships in long-term time series and palaeoecological records: deep sea as a test bed. Philos Trans R Soc Lond B Biol Sci. (2016) 371(1694): 20150282. https://doi.org/10.1098/rstb.2015.0282
[24] Kanarachos S, Mathew J, Fitzpatrick M E. Instantaneous vehicle fuel consumption estimation using smartphones and recurrent neural networks. Expert Systems with Application. (2019) 120(APR.): 436-447. https://doi.org/10.1016/j.eswa.2018.12.006
[25] Bieschewski S, Parcerisa J M, Antonio González. An Energy-Efficient Memory Unit for Clustered Microarchitectures. IEEE Transactions on Computers. (2016) 65(8): 2631-2637. https://doi.org/10.1109/TC.2015.2493518