Research And Implementation Of Key Technologies Based On HoloLens Augmented Reality

The key technologies of augmented reality systems include virtual reality fusion,three-dimensional registration,and real-time interaction.The core technology of augmented reality is the detection and location of the target object,also called three-dimensional registration.With the continuous development of big data and hardware technology,a large number of experiments have proved that the deep learning technology with convolutional neural network as the core greatly improves the efficiency and accuracy of image detection compared with the traditional machine learning method that artificially designs and extracts feature points,and widely used in all walks of life.Aiming at the accuracy and efficiency requirements and interaction requirements of key technologies in augmented reality systems,this paper studies and implements the following aspects:(1)In order to realize the universality of data,it is considered from the aspects of background,illumination,the size of object and shooting angle.For the common laboratory equipment,self-made data sets and calibration,and use data enhancement methods such as flip,zoom,crop,etc.to expand data sets and obtain better training fit,and use the network model of self-training to compare with Faster R-CNN network.According to the experimental results,a better target detection network is selected.(2)Based on the bounding box returned by the object detection model,a ray collision method is calculated with the calculating the boundary frame,and the physical location of the detection target is spatially located to provide spatial coordinates for the superimposed model,then evaluate the greatest detection model from the subjective and objective feelings.(3)Performing texture rendering of different methods on the 3D device model produced in the pre-production,designing the baking transplantation method to select the best result to adapt to the computing power of the augmented reality device.(4)Combine the above three points,and use the basic interaction method of the HoloLens such as Voice,Gesture and Gaze to design different interaction modes for the overlay model.(5)Using the HoloLens,which is an augmented reality device,designed and developed a detection and positioning interaction system based on the augmented reality device.The system can detect and locate the laboratory equipment and superimpose the model according to the spatial location.It can also perform various interactions on the model in real time,such as decomposition,movement,rotation,and so on.The above experimental results show that the method proposed in this paper can accurately detect the positioning laboratory equipment,and can have a simple,smooth and comfortable experience and interactive function of the overlay model,providing a new visualize interactive experience for users in equipment detection,assembly and disassembly learning.