期刊名称：International Journal of Electrical and Computer Engineering
电子版ISSN：2088-8708
出版年度：2018
卷号：8
期号：1
页码：483-496
DOI：10.11591/ijece.v8i1.pp483-496
语种：English
出版社：Institute of Advanced Engineering and Science (IAES)
摘要：Myoelectric pattern recognition (M-PR) is used to detect user’s intention to achieve a smooth interaction between human and machine. To improve the performance of MPR, this paper proposes an optimisation of RBF-ELM parameters using hybridization of PSO (particle swarm optimisation) and RBF-ELM (SRBF-ELM). Besides, it introduces the hybridization of PSO, wavelet, and RBF-ELM (SW-RBF-ELM) to anticipate the local optima that possibly occurs in PSO. These proposed systems are employed to classify finger movements on the amputees and able-bodied subjects using electromyography signals. The experimental results show that on the able-bodied subjects, the accuracy of SW-RBF-ELM and SRBF-ELM is 95.71 % and 95.53 %, respectively. The improvement of wavelet mutation on the amputees is more significant than that on the able-bodied subjects. On the amputees, the SW-RBF-ELM and SRBF-ELM achieved the average accuracy of 94.27 % and 92.55 %, respectively. The experimental results also show that SW-RBF-ELM achieved the accuracy that is better than other well-known classifiers such as support vector machine (SVM), linear discriminant analysis (LDA) and k-nearest neighbour (kNN).
其他摘要：Myoelectric pattern recognition (M-PR) is used to detect user’s intention to achieve a smooth interaction between human and machine. To improve the performance of MPR, this paper proposes an optimisation of RBF-ELM parameters using hybridization of PSO (particle swarm optimisation) and RBF-ELM (SRBF-ELM). Besides, it introduces the hybridization of PSO, wavelet, and RBF-ELM (SW-RBF-ELM) to anticipate the local optima that possibly occurs in PSO. These proposed systems are employed to classify finger movements on the amputees and able-bodied subjects using electromyography signals. The experimental results show that on the able-bodied subjects, the accuracy of SW-RBF-ELM and SRBF-ELM is 95.71 % and 95.53 %, respectively. The improvement of wavelet mutation on the amputees is more significant than that on the able-bodied subjects. On the amputees, the SW-RBF-ELM and SRBF-ELM achieved the average accuracy of 94.27 % and 92.55 %, respectively. The experimental results also show that SW-RBF-ELM achieved the accuracy that is better than other well-known classifiers such as support vector machine (SVM), linear discriminant analysis (LDA) and k-nearest neighbour (kNN).
关键词：Instrumentation and Control;Computer and Informatics;Classification; Pattern recognition; Electromyography; Extreme leraning machine; Wavelet
