Method and system for gesture recognition based on related characteristics of wrist tendon pressures

Abstract

The invention discloses a method and system for gesture recognition based on related characteristics of wrist tendon pressures. Complicated external equipment is not needed; and a user only needs to wear a bracelet equipped with a few of dual-row arranged pressure sensors on the wrist so as to achieve high-accuracy recognition of various gestures. On one side, convenience of the gesture recognition is enhanced greatly, and the method and system are applicable to various occasions. On the other side, the scheme is characterized in that optimization selection based on human body structures is conducted to placement positions of the pressure sensors; more effective characteristics of collected pressure information are extracted, and direct matching of pressure values is avoided; and a more advanced algorithm is selected, so that gesture recognition accuracy is greatly increased. In current experiments, the highest recognition accuracy of 24 gestures exceeds 95%.

Description

technical field [0001] The invention relates to the technical field of human-computer interaction, in particular to a gesture recognition method and system based on pressure-related characteristics of wrist tendons. Background technique [0002] Gesture recognition refers to the use of certain technologies to classify gesture types. At present, gesture recognition methods include video-based gesture recognition, that is, using a camera to obtain image information of the human hand, and then using image processing methods to extract the characteristics of different gestures for gesture recognition; there is also a gesture recognition based on electromyographic signals (EMG). Gesture recognition, that is, using the method of arranging myoelectric sensors on a large scale on the human arm to obtain the changes of myoelectric signals during different gestures, so as to classify gestures; in addition, there is also a gesture recognition based on ultrasound, which is passed Ultra...

AI Technical Summary

Problems solved by technology

[0004] The video-based gesture recognition solution needs to be used in a scene equipped with a camera, which greatly limits its scope of use. In addition, the effect of gesture recognition is also affected by conditions such as the lighting of the camera scene; The current gesture recognition scheme needs to arrange a large number of sensor electrodes on the forearm of the human body, which will seriously affect the user experience and greatly limit the application scenarios; the gesture recognition scheme based on ultrasound requires the use of a special ultrasonic probe and supporting processing equipment. The cost is high and requires professional technical support, which is not conducive to large-scale promotion

The gesture recognition scheme based on a large-scale pressure sensor array requires a large number of pressure sensors, so its cost will be extremely high. In addition, the method of directly matching the pressure value or pressure curve for gesture recognition will lead to low measurement accuracy.

Images