Method for controlling smart televisions via three-dimensional gestures on basis of binocular structured light

Abstract

The invention discloses a method for controlling smart televisions via three-dimensional gestures on basis of binocular structured light. The method comprises the following steps of: synchronously acquiring images through a binocular camera; reconstructing a three-dimensional image according to an obtained left-right view; carrying out preprocessing; recognizing a gesture action after three-dimensional gestures are segmented; and converting the recognized gesture action into an operation instruction for a smart television, and executing the operation instruction. According to the method, a binocular vision and structured light combined manner is adopted, so that the environmental anti-interference ability can be effectively improved, the influences, on fingertip positioning, of light conditions can be reduced, and the timeliness and effectiveness of smart televisions can be satisfied.

Description

technical field [0001] The invention relates to the field of human-computer interaction, in particular to a gesture control method. Background technique [0002] With the development of Internet technology, smart TV—a kind of open operating system and chip, has an open application platform, can realize two-way human-computer interaction function, integrates multiple functions such as audio-visual, entertainment, data, etc., to meet TV products with diversified and personalized needs of users have become the trend of TV. [0003] The traditional button-type remote control is mainly composed of an integrated circuit board and buttons used to generate different messages. There is a central processing unit CPU inside, and the CPU just inputs the password information of various menus of the TV set into it when the CPU is manufactured, and the remote control transmitter of the TV set can realize the remote control of the TV set as long as it sends out the corresponding password. ...

AI Technical Summary

Problems solved by technology

However, most of the current image processing technologies such as gesture detection and fingertip positioning are based on two-dimensional, which are suitable for close-range and simple backgrounds, and do not meet the interactive requirements of smart TVs.

And 3D gesture recognition can judge gesture changes in 3D space, and has attracted widespread attention in a more natural and intuitive way of interaction. However, at present, some 3D gesture recognition devices use data gloves. The naturalness of gesture interaction is lost, and the cost is also very expensive; some devices that use time-of-flight have high hardware requirements, high cost, and low resolution; devices based on binocular vision have a small viewing angle, in dim environments and feature The effect is poor when it is not obvious; and the device based on structured light is very easy to be affected by the light

It can be said that existing gesture recognition devices have defects in varying degrees.

Images