Behavior recognition method based on surface electromyography measurement of waist and shoulder

Abstract

The invention discloses a behavior recognition method based on surface electromyography measurement of the waist and shoulder. The method is to divide the motion process into time segments, set identification codes in the time segments, and compare the identification codes with preset values to determine whether the motion process conforms to the motion described by a preset curve so as to realizesignal recognition, the recognition method being realized by using a multi-physiological signal acquisition system. The recognition method can recognize the behavior of a person without motion capture under the premise that the degree of freedom of motion is limited, and does not require subjective attention and operation, is not long in time delay, does not occupy space, can realize the monitoring of the force exerting tendency, provides signal output in the fields of driving, entertainment interaction, public health monitoring, etc., which have low reliability requirements and frequent interactions, has broad application prospects, and can recognize the movement and also the change in the force exerting of a patient. This method can be used to monitor human lesions and assess labor risks, and has considerable practical value and economic potential.

Description

technical field [0001] The invention relates to the intersecting field of ergonomics and signal acquisition, in particular to a behavior recognition method based on waist and shoulder surface electromyography measurement. Background technique [0002] The measurement of myoelectricity is an important means of monitoring human body signals. It is relatively reliable in a laboratory environment and is widely used in medical and ergonomic research. The amplitude and frequency of muscle electric waves are detected by myoelectric sensors. The research on myoelectricity is quite mature, but the signal recognition reliability of myoelectricity itself is poor, and there are great differences among different objects. The software simulation of muscle and joint load is mainly based on the calculation of the muscle and bone dummy of inverse dynamics. The load time calculation is more accurate, but it is difficult to effectively identify the internal force when the movement is not obvio...

AI Technical Summary

Problems solved by technology

The research on myoelectricity is quite mature, but the recognition reliability of myoelectricity itself is poor, and there are great differences among different objects. The software simulation of muscle and joint load is mainly based on the calculation of the muscle and bone dummy of inverse dynamics. The load time calculation is more accurate, but it is difficult to effectively identify the internal force when the movement is not obvious

[0003] There are many deficiencies in the current EMG signal recognition: the preparation of EMG measurement requires many steps, the mechanism of muscle exertion is complex, personal measurement may involve ethical issues, the amplitude stability of the measured signal wave is poor, and it is easily affected by ECG and physical stress. The fat (ECG) and skin stretching interference of the watch have few precedents applied to the control system; it is not easy to apply to the field of automatic control with high reliability requirements, as well as labor conditions and entertainment scenes, and controls based on the electromyographic feedback signal The research mainly focuses on prosthetic control and human assistance. The human-machine contact interface in this kind of scene is stable in a small area, and the acquisition difficulty is small. It is rare to use the recognition of the muscle signals of the lower back to control the equipment, mainly because there is a lack of suitable and commercially valuable research scene

[0004] Human body signal output ability is limited by hand, foot and mouth, which often limits the interaction ability and efficiency of the audience, and cannot freely switch according to the needs of real scenes and interaction; human-computer interaction requires the time of human perception and output control, which will reduce human Efficiency of machine-computer interaction; people need to pay attention to output signals by using hands, mouth and feet, which affects thinking, and can not use subconscious behavior to help control; using subjective consciousness to achieve control will lead to a large load on the brain of the operation interaction, resulting in tension, fatigue, and lack of comfort. Redesigning a behavior recognition method based on electromyography measurement

Images