A DATASET AT A GLANCE...

Human gesture recognition based on Electromyography (EMG) is one of the most-appealing technology of today’s smart world. The electromyography signal that represent the electrical activity of the skeletal muscles, when used as a biometric trait finds varied applications such as: (i) To control the smart household and industrial appliances, (ii) To make the robots understand and respond back to the human movements; (iii) To provide an interactive and enjoyable gaming experience, (iv) To track and aid the patients under medical care and the people with movement disorders and (v) To assist the people in communication, who signal their need using hand gestures. However, the EMG signals always has noise associated with it because of bodily movements, sweating etc. Hence, research on improving EMG- based human gesture recognition is highly-demanded to support the aforesaid applications. If a researcher wants to improve an EMG- based biometric authentication procedure or perform biometric identification and subject-independent gesture recognition using electrode shift invariant techniques, he/ she might freely access the Gesture Recognition and Biometrics ElectroMyogram (GRABMyo) dataset from PhysioNet (https://physionet.org/content/grabmyo/1.1.0/). This EMG dataset has been acquired from about 43 healthy people, who are under the age between 24 and 35. In fact, the data collection involved a three-day identical experimental session to collect about 129 EMG recordings, which are consistent over time. However, the researcher has three citation requests to use this dataset and they are as follows:

• Jiang, N., Pradhan, A., & He, J. (2024). Gesture Recognition and Biometrics ElectroMyogram (GRABMyo) (version 1.1.0). PhysioNet. https://doi.org/10.13026/89dm-f662

• Pradhan, A., He, J. & Jiang, N. Multi-day dataset of forearm and wrist electromyogram for hand gesture recognition and biometrics. Sci Data 9, 733 (2022)

• Goldberger, A., Amaral, L., Glass, L., Hausdorff, J., Ivanov, P. C., Mark, R., ... & Stanley, H. E. (2000). PhysioBank, PhysioToolkit, and PhysioNet: Components of a new research resource for complex physiologic signals. Circulation [Online]. 101 (23), pp. e215–e220