Walking Gait Phase Detection Based on Acceleration Signals Using Voting-Weighted Integrated Neural Network
Human gait phase recognition is a significant technology for rehabilitation training robot, human disease diagnosis, artificial prosthesis, and so on. The efficient design of the recognition method for gait information is the key issue in the current gait phase division and eigenvalues extraction re...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2020-01-01
|
| Series: | Complexity |
| Online Access: | http://dx.doi.org/10.1155/2020/4760297 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832566270173642752 |
|---|---|
| author | Lei Yan Tao Zhen Jian-Lei Kong Lian-Ming Wang Xiao-Lei Zhou |
| author_facet | Lei Yan Tao Zhen Jian-Lei Kong Lian-Ming Wang Xiao-Lei Zhou |
| author_sort | Lei Yan |
| collection | DOAJ |
| description | Human gait phase recognition is a significant technology for rehabilitation training robot, human disease diagnosis, artificial prosthesis, and so on. The efficient design of the recognition method for gait information is the key issue in the current gait phase division and eigenvalues extraction research. In this paper, a novel voting-weighted integrated neural network (VWI-DNN) is proposed to detect different gait phases from multidimensional acceleration signals. More specifically, it first employs a gait information acquisition system to collect different IMU sensors data fixed on the human lower limb. Then, with dimensionality reduction and four-phase division preprocessing, key features are selected and merged as unified vectors to learn common and domain knowledge in time domain. Next, multiple refined DNNs are transferred to design a multistream integrated neural network, which utilizes the mixture-granularity information to exploit high-dimensional feature representative. Finally, a voting-weighted function is developed to fuse different submodels as a unified representation for distinguishing small discrepancy among different gait phases. The end-to-end implementation of the VWI-DNN model is fine-tuned by the loss optimization of gradient back-propagation. Experimental results demonstrate the outperforming performance of the proposed method with higher classification accuracy compared with the other methods, of which classification accuracy and macro-F1 is up to 99.5%. More discussions are provided to indicate the potential applications in combination with other works. |
| format | Article |
| id | doaj-art-3bee3cfa70c34fe285430f0f220dd321 |
| institution | Kabale University |
| issn | 1076-2787 1099-0526 |
| language | English |
| publishDate | 2020-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Complexity |
| spelling | doaj-art-3bee3cfa70c34fe285430f0f220dd3212025-02-03T01:04:39ZengWileyComplexity1076-27871099-05262020-01-01202010.1155/2020/47602974760297Walking Gait Phase Detection Based on Acceleration Signals Using Voting-Weighted Integrated Neural NetworkLei Yan0Tao Zhen1Jian-Lei Kong2Lian-Ming Wang3Xiao-Lei Zhou4Beijing Forestry University, Beijing 100083, ChinaBeijing Forestry University, Beijing 100083, ChinaBeijing Technology and Business University, Beijing 100048, ChinaBeijing Forestry University, Beijing 100083, ChinaBeijing Forestry University, Beijing 100083, ChinaHuman gait phase recognition is a significant technology for rehabilitation training robot, human disease diagnosis, artificial prosthesis, and so on. The efficient design of the recognition method for gait information is the key issue in the current gait phase division and eigenvalues extraction research. In this paper, a novel voting-weighted integrated neural network (VWI-DNN) is proposed to detect different gait phases from multidimensional acceleration signals. More specifically, it first employs a gait information acquisition system to collect different IMU sensors data fixed on the human lower limb. Then, with dimensionality reduction and four-phase division preprocessing, key features are selected and merged as unified vectors to learn common and domain knowledge in time domain. Next, multiple refined DNNs are transferred to design a multistream integrated neural network, which utilizes the mixture-granularity information to exploit high-dimensional feature representative. Finally, a voting-weighted function is developed to fuse different submodels as a unified representation for distinguishing small discrepancy among different gait phases. The end-to-end implementation of the VWI-DNN model is fine-tuned by the loss optimization of gradient back-propagation. Experimental results demonstrate the outperforming performance of the proposed method with higher classification accuracy compared with the other methods, of which classification accuracy and macro-F1 is up to 99.5%. More discussions are provided to indicate the potential applications in combination with other works.http://dx.doi.org/10.1155/2020/4760297 |
| spellingShingle | Lei Yan Tao Zhen Jian-Lei Kong Lian-Ming Wang Xiao-Lei Zhou Walking Gait Phase Detection Based on Acceleration Signals Using Voting-Weighted Integrated Neural Network Complexity |
| title | Walking Gait Phase Detection Based on Acceleration Signals Using Voting-Weighted Integrated Neural Network |
| title_full | Walking Gait Phase Detection Based on Acceleration Signals Using Voting-Weighted Integrated Neural Network |
| title_fullStr | Walking Gait Phase Detection Based on Acceleration Signals Using Voting-Weighted Integrated Neural Network |
| title_full_unstemmed | Walking Gait Phase Detection Based on Acceleration Signals Using Voting-Weighted Integrated Neural Network |
| title_short | Walking Gait Phase Detection Based on Acceleration Signals Using Voting-Weighted Integrated Neural Network |
| title_sort | walking gait phase detection based on acceleration signals using voting weighted integrated neural network |
| url | http://dx.doi.org/10.1155/2020/4760297 |
| work_keys_str_mv | AT leiyan walkinggaitphasedetectionbasedonaccelerationsignalsusingvotingweightedintegratedneuralnetwork AT taozhen walkinggaitphasedetectionbasedonaccelerationsignalsusingvotingweightedintegratedneuralnetwork AT jianleikong walkinggaitphasedetectionbasedonaccelerationsignalsusingvotingweightedintegratedneuralnetwork AT lianmingwang walkinggaitphasedetectionbasedonaccelerationsignalsusingvotingweightedintegratedneuralnetwork AT xiaoleizhou walkinggaitphasedetectionbasedonaccelerationsignalsusingvotingweightedintegratedneuralnetwork |