Hand gesture-based interface with multichannel sEMG band enabling unknown gesture discrimination

Tomczyński, Jakub; Kaczmarek, Piotr; Mańkowski, Tomasz

doi:10.1109/romoco.2015.7219713

Cited by 10 publications

(12 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…putEMG dataset consists of a series of trials that include execution of 8 hand gestures (see Figure 3). The gesture choice was based on former studies, active poses showing the highest discrimination [9,25,26] and best fitting for convenient sEMG-based HMI design were selected. The set consists of 7 active gestures — fist , flexion , extension , and 4 pinch gestures where the thumb meets one of the remaining fingers ( pinch index , pinch middle , pinch ring , pinch small —Figure 3e–h).…”

Section: Putemg Data Collectionmentioning

confidence: 99%

putEMG—A Surface Electromyography Hand Gesture Recognition Dataset

Kaczmarek

Mańkowski

Tomczyński

2019

Sensors

Self Cite

View full text Add to dashboard Cite

In this paper, we present a putEMG dataset intended for the evaluation of hand gesture recognition methods based on sEMG signal. The dataset was acquired for 44 able-bodied subjects and include 8 gestures (3 full hand gestures, 4 pinches and idle). It consists of uninterrupted recordings of 24 sEMG channels from the subject’s forearm, RGB video stream and depth camera images used for hand motion tracking. Moreover, exemplary processing scripts are also published. The putEMG dataset is available under a Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0). The dataset was validated regarding sEMG amplitudes and gesture recognition performance. The classification was performed using state-of-the-art classifiers and feature sets. An accuracy of 90% was achieved for SVM classifier utilising RMS feature and for LDA classifier using Hudgin’s and Du’s feature sets. Analysis of performance for particular gestures showed that LDA/Du combination has significantly higher accuracy for full hand gestures, while SVM/RMS performs better for pinch gestures. The presented dataset can be used as a benchmark for various classification methods, the evaluation of electrode localisation concepts, or the development of classification methods invariant to user-specific features or electrode displacement.

show abstract

Section: Putemg Data Collectionmentioning

confidence: 99%

putEMG—A Surface Electromyography Hand Gesture Recognition Dataset

Kaczmarek

Mańkowski

Tomczyński

2019

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…Although promising outcomes have been reported in well-controlled laboratory settings, the MPR approach has yet to be widely used in clinical settings. As many studies stated, some issues including novel motion interference (Scheme et al, 2011 ; Tomczyński et al, 2015 ; Ding et al, 2019 ), electrode shift (Hargrove et al, 2008 ; He and Zhu, 2017 ), or variation in force and limb orientation (Al-Timemy et al, 2015 ; Cheng et al, 2018 ; Nougarou et al, 2019 ) can significantly decay practical performance of the MPR systems. Among them, the novel motion interference is a critical issue for implementing a MPR system in real-world scenarios.…”

Section: Introductionmentioning

confidence: 99%

“…Many studies in this area have been conducted to alleviate the novel motion interference (Liu and Huang, 2009 ; Scheme et al, 2011 , 2013 ; Amsuess et al, 2015 ; Tomczyński et al, 2015 ; Ding et al, 2017 , 2019 ; Robertson et al, 2019 ). Generally speaking, common solutions to this issue can be explained to design a filtering scheme that could only pass through target motion patterns while rejecting any unlearned novel motion patterns.…”

Section: Introductionmentioning

confidence: 99%

“…Generally speaking, common solutions to this issue can be explained to design a filtering scheme that could only pass through target motion patterns while rejecting any unlearned novel motion patterns. The implementation concepts of these filtering schemes can be divided into two categories: one is based on confidence (Scheme et al, 2013 ; Tomczyński et al, 2015 ; Robertson et al, 2019 ) and the other uses a one-vs.-all classification rule (Liu and Huang, 2009 ; Scheme et al, 2011 ; Amsuess et al, 2015 ; Ding et al, 2017 , 2019 ). As representative examples of the confidence-based schemes, Scheme et al ( 2013 ) and Robertson et al ( 2019 ) utilized the classifier to provide a confidence score for each decision and subsequently rejected samples from novel motion tasks with a score below the threshold.…”

Section: Introductionmentioning

confidence: 99%

“…As representative examples of the confidence-based schemes, Scheme et al ( 2013 ) and Robertson et al ( 2019 ) utilized the classifier to provide a confidence score for each decision and subsequently rejected samples from novel motion tasks with a score below the threshold. Similarly, Tomczyński et al ( 2015 ) used the output entropy function of an artificial neural network to discriminate between target and novel motion tasks. The basic idea of the one-vs.-all classification rule is to model each known class, i.e., target motion task, by primarily learning from a training set containing only the samples of that class, and then a test sample will be rejected if it does not belong to any known class.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Rejecting Novel Motions in High-Density Myoelectric Pattern Recognition Using Hybrid Neural Networks

Chen

et al. 2022

Front. Neurorobot.

View full text Add to dashboard Cite

The objective of this study is to develop a method for alleviating a novel pattern interference toward achieving a robust myoelectric pattern-recognition control system. To this end, a framework was presented for surface electromyogram (sEMG) pattern classification and novelty detection using hybrid neural networks, i.e., a convolutional neural network (CNN) and autoencoder networks. In the framework, the CNN was first used to extract spatio-temporal information conveyed in the sEMG data recorded via high-density (HD) 2-dimensional electrode arrays. Given the target motion patterns well-characterized by the CNN, autoencoder networks were applied to learn variable correlation in the spatio-temporal information, where samples from any novel pattern appeared to be significantly different from those from target patterns. Therefore, it was straightforward to discriminate and then reject the novel motion interferences identified as untargeted and unlearned patterns. The performance of the proposed method was evaluated with HD-sEMG data recorded by two 8 × 6 electrode arrays placed over the forearm extensors and flexors of 9 subjects performing seven target motion tasks and six novel motion tasks. The proposed method achieved high accuracies over 95% for identifying and rejecting novel motion tasks, and it outperformed conventional methods with statistical significance (p < 0.05). The proposed method is demonstrated to be a promising solution for rejecting novel motion interferences, which are ubiquitous in myoelectric control. This study will enhance the robustness of the myoelectric control system against novelty interference.

show abstract