Hand Movement Activity-Based Character Input System on a Virtual Keyboard

Abstract

Nowadays, gesture-based technology is revolutionizing the world and lifestyles, and the users are comfortable and care about their needs, for example, in communication, information security, the convenience of day-to-day operations and so forth. In this case, hand movement information provides an alternative way for users to interact with people, machines or robots. Therefore, this paper presents a character input system using a virtual keyboard based on the analysis of hand movements. We analyzed the signals of the accelerometer, gyroscope, and electromyography (EMG) for movement activity. We explored potential features of removing noise from input signals through the wavelet denoising technique. The envelope spectrum is used for the analysis of the accelerometer and gyroscope and cepstrum for the EMG signal. Furthermore, the support vector machine (SVM) is used to train and detect the signal to perform character input. In order to validate the proposed model, signal information is obtained from predefined gestures, that is, “double-tap”, “hold-fist”, “wave-left”, “wave-right” and “spread-finger” of different respondents for different input actions such as “input a character”, “change character”, “delete a character”, “line break”, “space character”. The experimental results show the superiority of hand gesture recognition and accuracy of character input compared to state-of-the-art systems.

One-shot random forest model calibration for hand gesture decoding

“…Many myoelectric control systems use machine learning to map the electromyographic (EMG) signals [1][2][3][4][5] to control commands for human-machine interfaces, e.g. prosthesis [6][7][8][9][10][11][12] and virtual keyboards [13,14]. Most modern myoelectric control machine learning models require a large amount of data from a user to learn a bespoke and user-specific map [7,15,16].…”

One-Shot Random Forest Model Calibration for Hand Gesture Decoding

“…Many myoelectric control systems use machine learning to map the electromyographic (EMG) signals [1, 2, 3, 4, 5] to control commands for human-machine interfaces, e.g. prosthesis [6, 7, 8, 9, 10, 11, 12] and virtual keyboards [13, 14]. Most modern myoelectric control machine learning models require a large amount of data from a user to learn a bespoke and user-specific map [7, 15, 16].…”

An All-Fabric Tactile-Sensing Keypad with Uni-Modal and Ultrafast Response/Recovery Time for Smart Clothing Applications

“…Traditionally, keypads are made of solid-state materials and circuitry. With the rapid development of wearable applications, people are beginning to explore other forms of devices that can perform or replace keypad/keyboard functions, such as finger gesture recognition, − voice inputs, virtual keypad projection, , etc. However, these new forms of devices either do not give users the intuitive and unobstructed experience, or their accuracy and dexterity are still not up to the application requirements.…”