A Novel Deep Neural Network Method for HAR-Based Team Training Using Body-Worn Inertial Sensors
Human activity recognition (HAR) became a challenging issue in recent years. In this paper, we propose a novel approach to tackle indistinguishable activity recognition based on human wearable sensors. Generally speaking, vision-based solutions struggle with low illumination environments and partial...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/21/8507 |
| _version_ | 1797466446456946688 |
|---|---|
| author | Yun-Chieh Fan Yu-Hsuan Tseng Chih-Yu Wen |
| author_facet | Yun-Chieh Fan Yu-Hsuan Tseng Chih-Yu Wen |
| author_sort | Yun-Chieh Fan |
| collection | DOAJ |
| description | Human activity recognition (HAR) became a challenging issue in recent years. In this paper, we propose a novel approach to tackle indistinguishable activity recognition based on human wearable sensors. Generally speaking, vision-based solutions struggle with low illumination environments and partial occlusion problems. In contrast, wearable inertial sensors can tackle this problem and avoid revealing personal privacy. We address the issue by building a multistage deep neural network framework that interprets accelerometer, gyroscope, and magnetometer data that provide useful information of human activities. Initially, the stage of variational autoencoders (VAE) can extract the crucial information from raw data of inertial measurement units (IMUs). Furthermore, the stage of generative adversarial networks (GANs) can generate more realistic human activities. Finally, the transfer learning method is applied to enhance the performance of the target domain, which builds a robust and effective model to recognize human activities. |
| first_indexed | 2024-03-09T18:40:00Z |
| format | Article |
| id | doaj.art-4e4bd2dd1bc041e1bb1ec2312f6fbc2c |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T18:40:00Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-4e4bd2dd1bc041e1bb1ec2312f6fbc2c2023-11-24T06:49:07ZengMDPI AGSensors1424-82202022-11-012221850710.3390/s22218507A Novel Deep Neural Network Method for HAR-Based Team Training Using Body-Worn Inertial SensorsYun-Chieh Fan0Yu-Hsuan Tseng1Chih-Yu Wen2Simulator Systems Section, Aeronautical System Research Division, National Chung-Shan Institute of Science and Technology, Taichung 407, TaiwanDepartment of Computer Science and Engineering, National Chung Hsing University, Taichung 402, TaiwanDepartment of Electrical Engineering, National Chung Hsing University, Taichung 402, TaiwanHuman activity recognition (HAR) became a challenging issue in recent years. In this paper, we propose a novel approach to tackle indistinguishable activity recognition based on human wearable sensors. Generally speaking, vision-based solutions struggle with low illumination environments and partial occlusion problems. In contrast, wearable inertial sensors can tackle this problem and avoid revealing personal privacy. We address the issue by building a multistage deep neural network framework that interprets accelerometer, gyroscope, and magnetometer data that provide useful information of human activities. Initially, the stage of variational autoencoders (VAE) can extract the crucial information from raw data of inertial measurement units (IMUs). Furthermore, the stage of generative adversarial networks (GANs) can generate more realistic human activities. Finally, the transfer learning method is applied to enhance the performance of the target domain, which builds a robust and effective model to recognize human activities.https://www.mdpi.com/1424-8220/22/21/8507human activity recognitionvariational autoencodergenerative adversarial networks |
| spellingShingle | Yun-Chieh Fan Yu-Hsuan Tseng Chih-Yu Wen A Novel Deep Neural Network Method for HAR-Based Team Training Using Body-Worn Inertial Sensors Sensors human activity recognition variational autoencoder generative adversarial networks |
| title | A Novel Deep Neural Network Method for HAR-Based Team Training Using Body-Worn Inertial Sensors |
| title_full | A Novel Deep Neural Network Method for HAR-Based Team Training Using Body-Worn Inertial Sensors |
| title_fullStr | A Novel Deep Neural Network Method for HAR-Based Team Training Using Body-Worn Inertial Sensors |
| title_full_unstemmed | A Novel Deep Neural Network Method for HAR-Based Team Training Using Body-Worn Inertial Sensors |
| title_short | A Novel Deep Neural Network Method for HAR-Based Team Training Using Body-Worn Inertial Sensors |
| title_sort | novel deep neural network method for har based team training using body worn inertial sensors |
| topic | human activity recognition variational autoencoder generative adversarial networks |
| url | https://www.mdpi.com/1424-8220/22/21/8507 |
| work_keys_str_mv | AT yunchiehfan anoveldeepneuralnetworkmethodforharbasedteamtrainingusingbodyworninertialsensors AT yuhsuantseng anoveldeepneuralnetworkmethodforharbasedteamtrainingusingbodyworninertialsensors AT chihyuwen anoveldeepneuralnetworkmethodforharbasedteamtrainingusingbodyworninertialsensors AT yunchiehfan noveldeepneuralnetworkmethodforharbasedteamtrainingusingbodyworninertialsensors AT yuhsuantseng noveldeepneuralnetworkmethodforharbasedteamtrainingusingbodyworninertialsensors AT chihyuwen noveldeepneuralnetworkmethodforharbasedteamtrainingusingbodyworninertialsensors |