ABSTRACT
With the ever increasing elderly population, human activity trackers can help monitor the daily physical activities performed by the elderly in order to contribute towards improvements in independent living and quality of life.
Little activity recognition research has explored the use of multiple Inertial Measurement Units (IMU) to capture both simple and complex human activities. Therefore, existing research may not benefit the monitoring of the elderly population wherein the complexity in the details of changes are not captured. This work proposes a multi-sensor approach measuring acceleration and quaternion values to recognise both simple and complex daily living activities using a deep learning approach. We compare and evaluate the performance of using 1, 3 and 5 on-body IMU sensors to train CNN and LSTM networks with both acceleration and quaternion values. The results show that the adoption of the quaternion values from 5 on-body sensors using the LSTM model outperforms all other models (F1-score=0.9606). This high performance provides many opportunities for the accurate monitoring of complex daily living activities.
- Shamir Alavi, Dennis Arsenault, and Anthony Whitehead. 2016. Quaternion-Based Gesture Recognition Using Wireless Wearable Motion Capture Sensors. Sensors (Basel, Switzerland) 16, 5 (5 2016).Google Scholar
- Vidya Balu and Sasikumar P. 2022. Wearable Multi-Sensor Data Fusion Approach for Human Activity Recognition Using Machine Learning Algorithms. SSRN Electronic Journal (2 2022).Google Scholar
- R. S. Bucks, D. L. Ashworth, G. K. Wilcock, and K. Siegfried. 1996. Assessment of activities of daily living in dementia: development of the Bristol Activities of Daily Living Scale. Age and ageing 25, 2 (1996), 113--120.Google Scholar
- Lei Gao, Alan K. Bourke, and John Nelson. 2011. A system for activity recognition using multi-sensor fusion. In Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. IEEE, Boston, MA, USA, 7869--7872.Google Scholar
- Lei Gao, A. K. Bourke, and John Nelson. 2014. Evaluation of accelerometer based multi-sensor versus single-sensor activity recognition systems. Medical Engineering & Physics 36, 6 (6 2014), 779--785.Google Scholar
- Lei Gao, A. K. Bourke, and John Nelson. 2014. Evaluation of accelerometer based multi-sensor versus single-sensor activity recognition systems. Medical Engineering and Physics 36, 6 (2014), 779--785.Google ScholarCross Ref
- Mohammed Mehedi Hassan, Md Zia Uddin, Amr Mohamed, and Ahmad Almogren. 2018. A robust human activity recognition system using smartphone sensors and deep learning. Future Generation Computer Systems 81 (4 2018), 307--313.Google Scholar
- Sepp Hochreiter and Jürgen Schmidhuber. 1997. Long Short-Term Memory. Neural Computation 9, 8 (11 1997), 1735--1780.Google ScholarDigital Library
- Sebastian O.H. Madgwick, Andrew J.L. Harrison, and Ravi Vaidyanathan. 2011. Estimation of IMU and MARG orientation using a gradient descent algorithm. IEEE International Conference on Rehabilitation Robotics (2011).Google ScholarCross Ref
- Sakorn Mekruksavanich and Anuchit Jitpattanakul. 2021. Deep Convolutional Neural Network with RNNs for Complex Activity Recognition Using Wrist-Worn Wearable Sensor Data. Electronics 2021, Vol. 10, Page 1685 10, 14 (7 2021), 1685.Google ScholarCross Ref
- Zoran Milanović, Saša Pantelić, Nebojša Trajković, Goran Sporiš, Radmila Kostić, and Nic James. 2013. Age-related decrease in physical activity and functional fitness among elderly men and women. Clinical interventions in aging 8 (2013), 549--556.Google Scholar
- Vimala Nunavath, Sahand Johansen, Tommy Sandtorv Johannessen, Lei Jiao, Bjørge Herman Hansen, Sveinung Berntsen, and Morten Goodwin. 2021. Deep Learning for Classifying Physical Activities from Accelerometer Data. Sensors 2021, Vol. 21, Page 5564 21, 16 (8 2021), 5564.Google ScholarCross Ref
- Madhuri Panwar, S. Ram Dyuthi, K. Chandra Prakash, Dwaipayan Biswas, Amit Acharyya, Koushik Maharatna, Arvind Gautam, and Ganesh R. Naik. 2017. CNN based approach for activity recognition using a wrist-worn accelerometer. Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS (9 2017), 2438--2441.Google Scholar
- Charissa Ann Ronao and Sung Bae Cho. 2016. Human activity recognition with smartphone sensors using deep learning neural networks. Expert Systems with Applications 59 (10 2016), 235--244.Google Scholar
- Emilio Sansano, Raúl Montoliu, and Óscar Belmonte Fernández. 2020. A study of deep neural networks for human activity recognition. Computational Intelligence 36, 3 (3 2020).Google Scholar
- Meng Shang, Yiyuan Zhang, Ahmed Youssef Ali Amer, Ine D'Haeseleer, and Bart Vanrumste. 2021. Bathroom activities monitoring for older adults by a wrist-mounted accelerometer using a hybrid deep learning model. Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS (2021), 7112--7115.Google ScholarCross Ref
- Y. E. Shin, W. H. Choi, and T. M. Shin. 2014. Physical activity recognition based on rotated acceleration data using quaternion in sedentary behavior: A preliminary study. 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014 (11 2014), 4976--4978.Google Scholar
- Office for National Statistics. 2018. Living longer - Office for National Statistics. Technical Report.Google Scholar
- Wenchuan Wei, Keiko Kurita, Jilong Kuang, and Alex Gao. 2021. Real-Time Limb Motion Tracking with a Single IMU Sensor for Physical Therapy Exercises. Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS (11 2021), 7152--7157.Google ScholarCross Ref
- Kieran Woodward and Eiman Kanjo. 2020. iFidgetCube: Tangible Fidgeting Interfaces (TFIs) to Monitor and Improve Mental Wellbeing. IEEE Sensors Journal (2020).Google Scholar
- Kieran Woodward, Eiman Kanjo, David J. Brown, and T.M. McGinnity. 2021. Towards Personalised Mental Wellbeing Recognition On-Device using Transfer Learning "in the Wild". In IEEE International Smart Cities Conference 2021.Google Scholar
- Kieran Woodward, Eiman Kanjo, Andreas Oikonomou, and Alan Chamberlain. 2020. LabelSens: enabling real-time sensor data labelling at the point of collection using an artificial intelligence-based approach. Personal and Ubiquitous Computing 24, 5 (6 2020), 709--722.Google Scholar
- xsens. 2021. Xsens DOT User Manual. https://www.xsens.com/hubfs/Downloads/Manuals/XsensDOTUserManual.pdfGoogle Scholar
Index Terms
- A multi-sensor deep learning approach for complex daily living activity recognition
Recommendations
An Adaptive Rule-Based Approach to Classifying Activities of Daily Living
ICHI '15: Proceedings of the 2015 International Conference on Healthcare InformaticsThe need for a human activity recognition system arises when designing a "health smart home" which monitors its occupants to assess their health status. In this work, a rule-based system was constructed to classify the common activities of daily living ...
Transformer-Based Recognition of Activities of Daily Living from Wearable Sensor Data
iWOAR '22: Proceedings of the 7th International Workshop on Sensor-based Activity Recognition and Artificial IntelligenceSmart support systems for the recognition of Activities of Daily Living (ADLs) can help elderly people live independently for longer improving their standard of living. Many machine learning approaches have been proposed lately for Human Activity ...
Human Activity Behavioural Pattern Recognition in Smart Home with Long-Hour Data Collection
AbstractThe research on human activity recognition has provided novel solutions to many applications like health care, sports, and user profiling. Considering the complex nature of human activities, it is still challenging even after effective and ...
Comments