Design Of Electronic Distorting Mirror System Based On Kinect Somatosensory Interaction Technology

As we all know,the distorting mirror is a precious childhood memories of many people.The traditional optical distorting mirror was based on the principle of optical imaging.According to the irregular light reflection and focusing of the mirror,different images are presented.However,it has gradually faded people's attention due to some disadvantages including large footprint,single function point and no effect of human-computer interaction.On the other hand,with the improvement of people's spiritual and cultural level,technology visual entertainment is being more and more popular.In order to satisfy visual entertainment and reproduce the charm of the mirror,the concept of electronic mirror was introduced.And,it combined with modern technologies including computer application,video processing and virtual reality was used to collect video images and identify the movements of tourists in real time.According to the recognition result,the image were deformed,distorted and then displayed.The design of the electronic distorting mirror system based on Kinect somatosensory interaction technology was different from the traditional one.This system used Kinect somatosensory technology which combined the bone tracking and imaging technology with the image distortion technology to realize the local distorting effect instead of the traditional global distortion effect.Meanwhile,the gesture recognition triggers the distortion effect instead of the traditional time gradient switching effect,the entire system runs smoothly and lets players experience the gameplay of human-computer interaction.The main contents of this paper are as follows:(1)In the process of image distortion,the image overlap and void problem would occur in the forward mapping.In this paper,the inverse mapping and bilinear inter:polation algorithm were used to solve this problem by calculating the location of each pixel point of object image in the source image.On the basis of the overall distortion,in order to achieve the local distortion effect,this paper adopts the bone tracking technology,took the skeleton point as the center,selected the specified area and used the corresponding deformation function to realize the local distortion effect of the mirror.(2)In the method of realizing the character becoming short,some background deformation and vacancy problems are involved.This paper proposed the Kinect imaging algorithm technology,which acquired depth image and color image through Kinect camera,and used joint bilateral filtering algorithm to process depth image by filtering.And the automatic generation algorithm of the ternary graph was realized,then,the color image and the ternary graph were selected as the inputs in shared sample point algorithm.Next,the sample region expansion was enlarged,sample collection and local smoothing would realize the character extraction of image.Finally,the extracted character map was processed by scaling algorithm to achieve a short character effect.(3)In the character gesture recognition method,this paper proposed a corresponding method for simple gestures and complex gestures.For some simple gestures,item was defined by the position and angle relationship between different skeleton points.For some complex gestures,the gesture features were extracted to construct gesture feature vectors.And the library of gesture template was trained by Visual Gesture Builder.The dynamic regularization algorithm was applied for similarity match of test gesture and template gesture.The template gesture with the shortest distance was regarded as the recognition gesture.Meanwhile,the test environment was established to validate the effect of gesture recognition for verifying the feasibility of algorithm.