Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Masoud Moradi; Sina Shamekhi
Volume 16, Issue 2 , September 2022, , Pages 167-182
Abstract
In recent years, the fabrication of devices that can facilitate the difficulty of communication between deaf people and the general public and translate sign language has attracted interest from researchers. But problems such as low accuracy and calculation speed and the high cost of tools have hindered ...
Read More
In recent years, the fabrication of devices that can facilitate the difficulty of communication between deaf people and the general public and translate sign language has attracted interest from researchers. But problems such as low accuracy and calculation speed and the high cost of tools have hindered the commercialization of research. Another challenge in making a practical tool is the necessity of good performance of the methods in the perspective of training by leave-one-subject-out or in other words classifying the data of a new person. Therefore, in this article, an efficient method for detecting hand gestures with the purpose of sign language translation has been presented, so that while using a method with lower dimensions, better performance can be obtained in all kinds of training methods. In the proposed method, the features consisting of the mean absolute value, variance, root mean square, waveform length, kurtosis, and skewness have been extracted from the empirical wavelet transformation of the electromyogram and inertial signals. Then, by the ReliefF method, effective features have been selected and for the classification of hand gestures, a support vector machine classifier has been used. The accuracy percentages of the proposed method on the PSL database and DB2, DB3, DB5, and DB7 datasets of the NinaPro database, have been respectively obtained as follows: 99.31%, 97.11%, 96.58%, 96.12%, and 97.32% in the word-subject training approach, 99.78%, 97.22%, 95.46%, 97.23%, and 97.72% in the word-all-subject training approach, and 97.43%, 94.68%, 89.66%, 91.55%, and 94.81% in the leave-one-subject-out method.
Khosro Rezaee; Fardin Ghaderi; Hamed Taheri Gorji; Javad Haddadnia
Volume 14, Issue 3 , October 2020, , Pages 195-208
Abstract
In modern prostheses, accurate processing of surface electromyogram (sEMG) signals has a significant effect on optimal muscle control. Although these signals are useful for diagnosing neuromuscular diseases, controlling prosthetic devices and detecting hand movements, non-robustness of EMG signal-based ...
Read More
In modern prostheses, accurate processing of surface electromyogram (sEMG) signals has a significant effect on optimal muscle control. Although these signals are useful for diagnosing neuromuscular diseases, controlling prosthetic devices and detecting hand movements, non-robustness of EMG signal-based recognition will give rise to various movement disorders. In this paper, we present an optimal approach to classify EMG signals for hand gesture and movement recognition, whose purpose is to be used as an efficient method of diagnosing neuromuscular diseases, determining the type of treatment and physiotherapy. The main assumption of this study is to improve the accuracy of recognition and therefore, we proposed a novel hand gesture and movement recognition model consists of three steps: (1) EMG signal features extraction based on time-frequency domain and fractal dimension features; (2) feature selection by soft ensembling of three procedures in which includes two sample T-tests, entropy and common wrapper feature reduction, and (3) classification based on kernel parameters optimization of SVM classifier by using Gases Brownian Motion Optimization (GBMO) algorithm. Two UC2018 DualMyo and UCI datasets have been considered to evaluate the proposed model. The first dataset is used to classify eight hand gestures and the second dataset is employed for the classification of six types of movement. The experiment results and statistical tests reveal that the designed approach has desirable performance with an average accuracy of above 98% in both datasets. Contrary to similar methods that perform classifications in finite classes with high error rates, the integrated method has satisfactory accuracy, robustness and reliability. Not only the proposed method contributes to the design of prostheses, but also provides effective outcomes for rehabilitation applications and clinical diagnosis processes.