sEMG-Based Identification of Hand Motion Commands Using Wavelet Neural Network Combined With Discrete Wavelet Transform

Duan, Feng; Dai, Lili; Chang, Wennan; Chen, Zengqiang; Zhu, Chi; Li, Wei

doi:10.1109/tie.2015.2497212

Cited by 161 publications

(78 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…TD feature extraction refers to procedures that extract properties from the EMG time series in its original form. These are commonly employed features in myoelectric pattern recognition systems due to their high accuracy in low-noise environments and low computational complexity, and provide intuitive information about muscle motor unit recruitment [30,[52][53][54][55][56]. The mean absolute value (MAV) feature or root mean square (RMS) feature, for example, is a TD feature that represents the average energy of the EMG signal within a window [57,58].…”

Section: Feature Extractionmentioning

confidence: 99%

Current Trends and Confounding Factors in Myoelectric Control: Limb Position and Contraction Intensity

Campbell

Phinyomark

Scheme

2020

Sensors

View full text Add to dashboard Cite

This manuscript presents a hybrid study of a comprehensive review and a systematic (research) analysis. Myoelectric control is the cornerstone of many assistive technologies used in clinical practice, such as prosthetics and orthoses, and human-computer interaction, such as virtual reality control. Although the classification accuracy of such devices exceeds 90% in a controlled laboratory setting, myoelectric devices still face challenges in robustness to variability of daily living conditions. The intrinsic physiological mechanisms limiting practical implementations of myoelectric devices were explored: the limb position effect and the contraction intensity effect. The degradation of electromyography (EMG) pattern recognition in the presence of these factors was demonstrated on six datasets, where classification performance was 13% and 20% lower than the controlled setting for the limb position and contraction intensity effect, respectively. The experimental designs of limb position and contraction intensity literature were surveyed. Current state-of-the-art training strategies and robust algorithms for both effects were compiled and presented. Recommendations for future limb position effect studies include: the collection protocol providing exemplars of at least 6 positions (four limb positions and three forearm orientations), three-dimensional space experimental designs, transfer learning approaches, and multi-modal sensor configurations. Recommendations for future contraction intensity effect studies include: the collection of dynamic contractions, nonlinear complexity features, and proportional control.

show abstract

Section: Feature Extractionmentioning

confidence: 99%

Current Trends and Confounding Factors in Myoelectric Control: Limb Position and Contraction Intensity

Campbell

Phinyomark

Scheme

2020

Sensors

View full text Add to dashboard Cite

show abstract

“…In the pattern recognition-based control approach, a classifier trained with supervised learning was employed to map sEMG activity to one of the predefined classes that correspond to different control commands. In the past decades, many methods have been proposed to design a sEMG pattern recognition-based interface, some of which have achieved high accuracy with many classes in a laboratory environment [2][3][4].…”

Section: Introductionmentioning

confidence: 99%

High-Density Surface EMG-Based Gesture Recognition Using a 3D Convolutional Neural Network

Chen

Zhang

et al. 2020

Sensors

View full text Add to dashboard Cite

High-density surface electromyography (HD-sEMG) and deep learning technology are becoming increasingly used in gesture recognition. Based on electrode grid data, information can be extracted in the form of images that are generated with instant values of multi-channel sEMG signals. In previous studies, image-based, two-dimensional convolutional neural networks (2D CNNs) have been applied in order to recognize patterns in the electrical activity of muscles from an instantaneous image. However, 2D CNNs with 2D kernels are unable to handle a sequence of images that carry information concerning how the instantaneous image evolves with time. This paper presents a 3D CNN with 3D kernels to capture both spatial and temporal structures from sequential sEMG images and investigates its performance on HD-sEMG-based gesture recognition in comparison to the 2D CNN. Extensive experiments were carried out on two benchmark datasets (i.e., CapgMyo DB-a and CSL-HDEMG). The results show that, where the same network architecture is used, 3D CNN can achieve a better performance than 2D CNN, especially for CSL-HDEMG, which contains the dynamic part of finger movement. For CapgMyo DB-a, the accuracy of 3D CNN was 1% higher than 2D CNN when the recognition window length was equal to 40 ms, and was 1.5% higher when equal to 150 ms. For CSL-HDEMG, the accuracies of 3D CNN were 15.3% and 18.6% higher than 2D CNN when the window length was equal to 40 ms and 150 ms, respectively. Furthermore, 3D CNN achieves a competitive performance in comparison to the baseline methods.

show abstract

“…This is the most distinctive characteristic of the proposed method compared with other works. The proposed wavelet has low power consumption and can be used for applications in which the amount of power consumption is crucial . This new wavelet function is realized by using FB method and needs no multiplier.…”

Section: Introductionmentioning

confidence: 99%

“…Because of its unique features, wavelet transform has very wide applications such as signal analysis, image and biomedical signals processing, and circuit analysis. [1][2][3][4][5][6][7][8][9][10][11][12][13] One of these brilliant features is the existence of a large number of different mother wavelets such as Haar, Meyer, Mallat, Daubechies, and Symlet, to name but a few. Each of these basic functions has its own characteristics that make it suitable for specific application.…”

Section: Introductionmentioning

confidence: 99%

A new discrete wavelet transform appropriate for hardware implementation

Meshkat

Dehghani

2020

Circuit Theory & Apps

View full text Add to dashboard Cite

Summary A new method for computation of discrete wavelet transform is introduced. The impulse response of the finite impulse response (FIR) filter, as the main block in filter bank method, is realized such that orthonormal wavelet transform is achieved without using any multiplier as the most challenging part of the FIR filters. It is proved that the calculated Sobolev regularity of the proposed wavelet is higher than that of Daubechies for the same support width. Compared with most popular known wavelets, simulation results of the newly introduced wavelet for signal denoising prove remarkable advantage of our wavelet especially in terms of complexity and power consumption. The proposed wavelet and well‐known Daubechies wavelet are separately implemented on the Spartan‐6 FPGA (Field Programmable Gate Array) to compare the performance of them. The occupied slices and LUTs (Lookup Table) in realization of the proposed wavelet in comparison with those of Daubechies wavelet are decreased by 50% and 56%, respectively. Its power consumption at 100 MHz is also reduced by 86% in comparison with the Daubechies wavelet, and maximum frequency is increased by 186%. Implementation with standard cells of a 0.18‐μm CMOS (Complementary Metal Oxide Semiconductor Field Effect Transistor) technology shows 75% and 85% reduction in the power and chip area, respectively.

show abstract

sEMG-Based Identification of Hand Motion Commands Using Wavelet Neural Network Combined With Discrete Wavelet Transform

Cited by 161 publications

References 55 publications

Current Trends and Confounding Factors in Myoelectric Control: Limb Position and Contraction Intensity

Current Trends and Confounding Factors in Myoelectric Control: Limb Position and Contraction Intensity

High-Density Surface EMG-Based Gesture Recognition Using a 3D Convolutional Neural Network

A new discrete wavelet transform appropriate for hardware implementation

Contact Info

Product

Resources

About