Research On Gestures Interactive Modes Of Three-Dimensional Display

At present,display device interaction is usually realized by simple keyboard,mouse,touch screen and other traditional ways.These interactive methods are simple and easy to operate.However,these interactive methods are somewhat unitary and lack of flexibility and visual effect with the popularization of 3D display devices.this paper proposes a non-contact interactive method in 3D display scene to adapt to the development trend of 3D display.This article mainly discusses the appropriate interaction mode in the field of three-dimensional display,analyzes the interaction habits according to three-dimensional display circumstances and select specific interactive gestures for researching gesture interaction algorithm in this article.First,a selected gesture is acquired by a Kinect somatosensory device.Second,the gesture is identified by an algorithm proposed in this paper.Then,interaction deviece provide a corresponding instruction interface to control the motion of the three-dimensional display device according to the recognition result of the algorithm.So,this paper achieves the ideal interaction of three-dimensional display devices.The specific research content of this paper is as follows:(1)A gesture interaction method based on 3D display is proposed,and interactive gestures based on 3D display application scenarios are also presented.In addition,we studied the deep image acquisition technology based on Kinect somatosensory device and the acquisition method of skeleton information.(2)Research correlation recognition algorithm in the process of static gesture recognition.It includes the acquisition of deep images.and separate hand area from the data of the Kinect human hand skeleton,and then the threshold is segmented by the gray distribution of the depth image to separate the isolated hand area,and the complete gesture is separated from background for subsequent processing.Then,the paper identify Corresponding gesture according to static algorithm.The core algorithms include the edge detection algorithm,the palm location algorithm,the curvature fingertip detection algorithm and the defect point elimination algorithm.(3)Research related algorithms of dynamic gesture recognition.It includes the location and tracking of human hand spatial location through the human hand skeleton data and the normalization of the related bone data so as to extract the gesture features without user difference.Then,the vector description of the gesture features is constructed,and the DTW algorithm is detected based on the starting point and the end point of the dynamic gesture.At the end of this chapter,the traditional DTW algorithm is improved(4)The involved algorithm program is transplanted into an existing three-dimensional display device,thereby realizing some basic operations of controlling the interaction of the three-dimensional display device through gestures,and verifying the pros and cons of the gesture recognition algorithm by observing whether the gesture instruction is valid.