Neural network algorithm-based fall detection modelling
Fall is a major threat among elderly people which may lead to injuries or even death. High recognition of developed fall detection model is very significance for the elderly to detect the falls. Related algorithm for the fall detection has been discussed in depth by researcher from the previous rese...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Penerbit UTHM
2020
|
| Subjects: | |
| Online Access: | http://eprints.uthm.edu.my/6113/1/AJ%202020%20%28186%29.pdf |
| _version_ | 1825710130907316224 |
|---|---|
| author | Mohd Yusoff, Ainul Husna Koh, Cheng Zhi Ngadimon, Khairulnizam Md Salleh, Salihatun |
| author_facet | Mohd Yusoff, Ainul Husna Koh, Cheng Zhi Ngadimon, Khairulnizam Md Salleh, Salihatun |
| author_sort | Mohd Yusoff, Ainul Husna |
| collection | UTHM |
| description | Fall is a major threat among elderly people which may lead to injuries or even death. High recognition of developed fall detection model is very significance for the elderly to detect the falls. Related algorithm for the fall detection has been discussed in depth by researcher from the previous research. However, the improvement of model accuracy is still needed. This article presents results of modelling for fall detection system by using nonlinear autoregression neural network NARnet algorithm. The algorithm is trained by network training function; LM, SCG and RP by collocation with threshold-based setting value. Two participants involved in obtaining the acceleration and angular velocity. The type of input source is divided into 4 different types for training. The selection of the model was based on the comparison of optimization epochs, magnitude of validate error or mean square error (MSE), magnitude of correlation performance, the convergence graph in term of MSE performance, accuracy of regression and non-zero value of autocorrelation graph. The simulated result shows that the training model of Type 2 is the best model with a training result of 6.1551mse, 40 epochs, time 0.06s, and 0.92742 accuracy. The result indicates that LM function produce the better solution when compared to another optimization function. In fact, the model accuracy demonstrated that the proposed method was reliable and efficient. |
| first_indexed | 2024-03-05T21:52:57Z |
| format | Article |
| id | uthm.eprints-6113 |
| institution | Universiti Tun Hussein Onn Malaysia |
| language | English |
| last_indexed | 2024-03-05T21:52:57Z |
| publishDate | 2020 |
| publisher | Penerbit UTHM |
| record_format | dspace |
| spelling | uthm.eprints-61132022-01-26T07:44:57Z http://eprints.uthm.edu.my/6113/ Neural network algorithm-based fall detection modelling Mohd Yusoff, Ainul Husna Koh, Cheng Zhi Ngadimon, Khairulnizam Md Salleh, Salihatun TK7800-8360 Electronics Fall is a major threat among elderly people which may lead to injuries or even death. High recognition of developed fall detection model is very significance for the elderly to detect the falls. Related algorithm for the fall detection has been discussed in depth by researcher from the previous research. However, the improvement of model accuracy is still needed. This article presents results of modelling for fall detection system by using nonlinear autoregression neural network NARnet algorithm. The algorithm is trained by network training function; LM, SCG and RP by collocation with threshold-based setting value. Two participants involved in obtaining the acceleration and angular velocity. The type of input source is divided into 4 different types for training. The selection of the model was based on the comparison of optimization epochs, magnitude of validate error or mean square error (MSE), magnitude of correlation performance, the convergence graph in term of MSE performance, accuracy of regression and non-zero value of autocorrelation graph. The simulated result shows that the training model of Type 2 is the best model with a training result of 6.1551mse, 40 epochs, time 0.06s, and 0.92742 accuracy. The result indicates that LM function produce the better solution when compared to another optimization function. In fact, the model accuracy demonstrated that the proposed method was reliable and efficient. Penerbit UTHM 2020 Article PeerReviewed text en http://eprints.uthm.edu.my/6113/1/AJ%202020%20%28186%29.pdf Mohd Yusoff, Ainul Husna and Koh, Cheng Zhi and Ngadimon, Khairulnizam and Md Salleh, Salihatun (2020) Neural network algorithm-based fall detection modelling. The International Journal of Integrated Engineering, 12 (3). pp. 138-150. ISSN 2229-838X https://doi.org/10.30880/ijie.2020.12.03.018 |
| spellingShingle | TK7800-8360 Electronics Mohd Yusoff, Ainul Husna Koh, Cheng Zhi Ngadimon, Khairulnizam Md Salleh, Salihatun Neural network algorithm-based fall detection modelling |
| title | Neural network algorithm-based fall detection modelling |
| title_full | Neural network algorithm-based fall detection modelling |
| title_fullStr | Neural network algorithm-based fall detection modelling |
| title_full_unstemmed | Neural network algorithm-based fall detection modelling |
| title_short | Neural network algorithm-based fall detection modelling |
| title_sort | neural network algorithm based fall detection modelling |
| topic | TK7800-8360 Electronics |
| url | http://eprints.uthm.edu.my/6113/1/AJ%202020%20%28186%29.pdf |
| work_keys_str_mv | AT mohdyusoffainulhusna neuralnetworkalgorithmbasedfalldetectionmodelling AT kohchengzhi neuralnetworkalgorithmbasedfalldetectionmodelling AT ngadimonkhairulnizam neuralnetworkalgorithmbasedfalldetectionmodelling AT mdsallehsalihatun neuralnetworkalgorithmbasedfalldetectionmodelling |