Biological signal gesture recognition device and method

Abstract

The invention relates to a biological signal gesture recognition device. The biological signal gesture recognition device comprises a power module, a muscle electric signal sensor module, a signal preprocessing module, an acceleration sensor and gyroscope module, a calculation and control module, a wireless module and a microcontroller. The muscle electric signal sensor module is used for picking up muscle electric signals of multiple muscle group surfaces, the signal preprocessing module is used for preprocessing the picked muscle electric signals, and the acceleration sensor and gyroscope module is used for sampling to obtain movement state information of the arm of a user. The calculation and control module is used for performing feature extraction and recognition according to the muscle electric signals and the movement state information of the arm. The wireless module is used for wirelessly transmitting results of feature extraction and recognition to controlled equipment. The invention further relates to a biological signal gesture recognition method. Multiple channel sensor signals can be processed at the same time, the size is small, power consumption is low, recognition precision is high, and speed is high.

Description

technical field [0001] The invention relates to a biological signal gesture recognition device and method. Background technique [0002] Since somatosensory devices have been used as human-computer interaction, the gesture-based interaction method has been deeply studied and applied to various fields because it is more intuitive and expresses diverse and flexible information. [0003] The human motor system is composed of a large number of bones, skeletal muscles and other connecting tissues, and the muscles contain a large number of muscle fibers. Muscle fibers will be stimulated by electrical signals from nearby motor neurons to perform contraction or stretching actions, and cooperate with tendons and bones to achieve the effect of joint movement. The electrical signals released by neurons to stimulate muscles are called electrical muscle signals (EMG), which can reflect the action state of muscles to a certain extent. [0004] When a person makes finger gestures and wri...

AI Technical Summary

Problems solved by technology

Existing gesture recognition devices mainly use the principles of image acquisition, infrared sensing, and acceleration sensing. The main disadvantages are fixed use scenarios, complex calculations, high energy consumption, and large volume.

This type of device is not suitable for mobile or unfixed application scenarios. In addition, considering the large size and power consumption, this type of device is usually used on the desktop, and cannot be miniaturized and wearable.

Images