Based On Gesture Recognition For Research Of Interactive Holographic Display

With the development of three-dimensional display technology,emerging display interfaces gradually enter people's lives,such as virtual reality(VR)and augmented reality(AR).holographic display as a true three-dimensional display technology is the ultimate goal of future display interfaces.Combining vision-based gesture recognition with a holographic display interface can bring users into an immersive scene experience,which is closer to the way people interact in nature and has farreaching research significance.The ability to recognize gestures in an interactive holographic display system is related to the success,comfort,and natural interaction of users.Therefore,seeking a stable and efficient recognition method is the focus of current research.This article mainly conducts in-depth research on static gesture and dynamic gesture recognition technologies;in addition,it studies the automatic control of spatial light modulators,builds and implements an interactive holographic display system based on static gestures.The main research work of this article is as foll ows:1.For static gesture recognition,accurate gesture segmentation is an important guarantee for subsequent recognition accuracy.In complex backgrounds,it is often difficult to accurately segment the gesture area.In this paper,according to the different clustering characteristics of skin color in HSV and YCr Cb color space Choose the method of fusing the skin color segmentation results of HSV and YCr Cb color space,that is,take and operate,which reduces the misjudgment of the background area and obtains a better segmentation effect than a single color space.Secondly,in the definition of gestures,the method of judging gestures based only on the angle between the fingertip and the centroid is easy to recognize errors due to the small sliding angle.Accurate recognition of stop,right swipe and left swipe gestures.2.For interactive holographic display: firstly,an interactive holographic display system is built;secondly,in order to realize automatic control of the holographic display in the interactive system,this article uses the spatial light modulator as an external monitor and uses the Psychopy toolbox for external monitoring The device presents the function of stimulus,which realizes the automatic control of the spatial light modulator;in this paper,the three-dimensional objects with 12881 viewpoints are dispersed at 1 ° intervals into 360 viewing angles,generating 360 holograms,and mapping static gesture commands into holograms The different reading order of the image finally realizes the interactive display of the left and right rotation of the holographic image under static gestures.3.For dynamic gesture recognition: This paper proposes a combined model based on 2D CNN + two-way LSTM network for dynamic gesture recognition.First,the 2D CNN is used to extract the feature vector of each image in the image sequence to form a feature vector sequence.The vector sequence is used as the input of the bidirectional LSTM network to extract the connection of dynamic gesture features in the tim e domain.The neural network model was tested using the Cambridge-Gesture public data set and obtained a higher recognition accuracy rate than other methods in the past,with an accuracy rate of 95.56%.