Multi-channel physiological signal somatosensory posture recognition method based on GS-LSSVM

Abstract

The invention discloses a multi-channel physiological signal somatosensory posture recognition method based on a GS-LSSVM. The method comprises the steps: 1, collecting a human body physiological signal original sample; 2, performing noise reduction through a wavelet threshold method; 3, extracting the physiological signal features; 4, establishing a multi-channel physiological signal somatosensory posture LSSVM recognition model; 5, carrying out PSO optimization training on a multi-channel physiological signal somatosensory posture SVM recognition model; 6, carrying out somatosensory posturerecognition model test based on the optimal multichannel physiological signals. The multi-channel physiological signal somatosensory posture recognition method based on the GS-LSSVM is high in recognition precision, good in real-time performance and good in recognition robustness.

Description

technical field [0001] The invention belongs to the field of gesture recognition, in particular to a GS-LSSVM-based multi-channel physiological signal somatosensory gesture recognition method. Background technique [0002] In the past two decades, the rapid development of the Internet and communication technology has laid the foundation for the development of the current Internet of Things. The subsequent rise of smart phones, wearable smart devices, smart cars, and virtual reality devices have gradually integrated into people's work and life. Everywhere. Traditional keyboard, mouse input methods and multi-touch technology can no longer fully meet people's needs, and human-computer interaction has once again ushered in new challenges. Among them, somatosensory gesture recognition technology is an important part of human-computer interaction technology, but it is also one of the difficulties in research. [0003] The traditional method uses image-based somatosensory gesture...

AI Technical Summary

Problems solved by technology

[0003] The traditional method uses image-based somatosensory gesture recognition technology. However, this method is easily affected by various factors such as lighting conditions and occlusions, and cannot fully meet the needs of somatosensory gesture recognition.

Images