Composite gesture recognition device and control method based on ultrasound and vision

Abstract

The invention discloses a composite gesture recognition device based on ultrasound and vision and a control method thereof. The device includes a visual recognition unit, an ultrasonic recognition unit, and a device for controlling the distance between the visual recognition unit and the ultrasonic recognition unit according to usage conditions. Carry out selective switching control unit, when the device is in the standby state, switch to select the ultrasonic recognition unit to recognize the gesture command, so that the device enters the working state; when the device is in the working state, if it is judged that the relative displacement between the gesture object and the device is greater than the predetermined value S, switch to select the visual recognition unit to recognize the gesture object, otherwise switch to select the ultrasonic recognition unit to recognize the gesture object. This solution can combine the advantages of camera recognition gestures and sound wave recognition gestures, and adopt different gesture recognition methods in different application environments, thereby appropriately increasing the standby time of multimedia devices, improving recognition accuracy and recognition gesture types, and widening the scope of application. greatly increase.

Description

technical field [0001] The invention relates to a gesture recognition device and a control method thereof. Background technique [0002] With the development of sensor technology and the maturity of various technologies, gesture recognition technology has entered the usability stage and has become a key symbol of high-end flagship smartphones, media tablets and smart glasses. Gesture recognition is to use various sensors to continuously collect the shape and displacement of hands / hand tools, complete modeling at intervals, form a sequence frame of model information, and then convert these information sequences into corresponding instructions for use Control implements certain operations. At present, various products and solutions have also begun to emerge. Most of the common gesture recognition on the market uses cameras on multimedia devices to recognize gestures. The technical process of gesture recognition based on cameras includes gesture image acquisition, gesture segm...

AI Technical Summary

Problems solved by technology

[0003] There are still some deficiencies in camera-based gesture recognition technology. For example, the palm is required to be basically parallel to the camera, causing users to feel uncomfortable when bending their hands for a long time; it mainly uses area changes to measure the distance between the palm and the camera. The error is large when the distance is long; in portable multimedia devices, camera-based gesture recognition technology consumes a lot of power, seriously affecting the standby time of portable multimedia devices, which is a fatal shortcoming for portable multimedia devices

Similarly, the disadvantage of gesture recognition technology based on sound waves is: since the palm is a surface, the sound wave is a spherical wave, and the sound wave returns to the sound wave receiver after being refracted by the palm, and only the signal with stronger energy is processed and processed after the signal processing. Analysis, cannot identify all signals reflected back through the palm

Therefore, it can only be recognized when the hand is displaced relative to the multimedia device. When the palm moves parallel to the multimedia direction for a small distance, the movement distance cannot be recognized; when the palm movement distance is small or there is no displacement, the finger movement cannot be recognized

Images