A Coupled Piezoelectric Sensor for MMG-Based Human-Machine Interfaces

Mechanomyography (MMG) is a technique of recording muscles activity that may be considered a suitable choice for human–machine interfaces (HMI). The design of sensors used for MMG and their spatial distribution are among the deciding factors behind their successful implementation to HMI. We present a new design of a MMG sensor, which consists of two coupled piezoelectric discs in a single housing. The sensor’s functionality was verified in two experimental setups related to typical MMG applications: an estimation of the force/MMG relationship under static conditions and a neural network-based gesture classification. The results showed exponential relationships between acquired MMG and exerted force (for up to 60% of the maximal voluntary contraction) alongside good classification accuracy (94.3%) of eight hand motions based on MMG from a single-site acquisition at the forearm. The simplification of the MMG-based HMI interface in terms of spatial arrangement is rendered possible with the designed sensor.

3D Skeletal Joints-Based Hand Gesture Spotting and Classification

This paper presents a novel approach to continuous dynamic hand gesture recognition. Our approach contains two main modules: gesture spotting and gesture classification. Firstly, the gesture spotting module pre-segments the video sequence with continuous gestures into isolated gestures. Secondly, the gesture classification module identifies the segmented gestures. In the gesture spotting module, the motion of the hand palm and fingers are fed into the Bidirectional Long Short-Term Memory (Bi-LSTM) network for gesture spotting. In the gesture classification module, three residual 3D Convolution Neural Networks based on ResNet architectures (3D_ResNet) and one Long Short-Term Memory (LSTM) network are combined to efficiently utilize the multiple data channels such as RGB, Optical Flow, Depth, and 3D positions of key joints. The promising performance of our approach is obtained through experiments conducted on three public datasets—Chalearn LAP ConGD dataset, 20BN-Jester, and NVIDIA Dynamic Hand gesture Dataset. Our approach outperforms the state-of-the-art methods on the Chalearn LAP ConGD dataset.

Review on EMG Acquisition and Classification Techniques: Towards Zero Retraining in the Influence of User and Arm Position Independence

The surface electromyogram (EMG) is widely studied and applied in machine control. Recent methods of classifying hand gestures reported classification rates of over 95%. However, the majority of the studies made were performed on a single user, focusing solely on the gesture classification. These studies are restrictive in practical sense: either focusing on just gestures, multi-user compatibility, or rotation independence. The variations in EMG signals due to these conditions present a challenge to the practical application of EMG devices, often requiring repetitious training per application. To the best of our knowledge, there is little comprehensive review of works done in EMG classification in the combined influence of user-independence, rotation and hand exchange. Therefore, in this paper we present a review of works related to the practical issues of EMG with a focus on the EMG placement, and recent acquisition and computing techniques to reduce training. First, we provided an overview of existing electrode placement schemes. Secondly, we compared the techniques and results of single-subject against multi-subject, multi-position settings. As a conclusion, the study of EMG classification in this direction is relatively new. However the results are encouraging and strongly indicate that EMG classification in a broad range of people and tolerance towards arm orientation is possible, and can pave way for more flexible EMG devices.

A Machine Learning Processing Pipeline for Reliable Hand Gesture Classification of FMG Signals with Stochastic Variance

ForceMyography (FMG) is an emerging competitor to surface ElectroMyography (sEMG) for hand gesture recognition. Most of the state-of-the-art research in this area explores different machine learning algorithms or feature engineering to improve hand gesture recognition performance. This paper proposes a novel signal processing pipeline employing a manifold learning method to produce a robust signal representation to boost hand gesture classifiers’ performance. We tested this approach on an FMG dataset collected from nine participants in 3 different data collection sessions with short delays between each. For each participant’s data, the proposed pipeline was applied, and then different classification algorithms were used to evaluate the effect of the pipeline compared to raw FMG signals in hand gesture classification. The results show that incorporating the proposed pipeline reduced variance within the same gesture data and notably maximized variance between different gestures, allowing improved robustness of hand gestures classification performance and consistency across time. On top of that, the pipeline improved the classification accuracy consistently regardless of different classifiers, gaining an average of 5% accuracy improvement.