Research On Mobile Augmented Reality Based Recognition Tracking Technology

In recent years,WebAR technology in mobile augmented reality has gained more and more attention and application in the fields of education,medical,tourism,engineering and film,etc.Users can use augmented reality applications by visiting web pages,which can bring a good interactive experience to users.However,in practical application,WebAR system is easily affected by objective factors such as lighting,photo angle,photo distance,size of object to be recognized,etc.The accuracy of target recognition,recognition speed and stability and real-time of natural tracking are difficult to be guaranteed,and it is impossible to provide stable and high quality service,and how to improve the use efficiency of network AR technology is an urgent problem to be solved.This article is developed to solve the above problems.(1)Because WebAR is easily affected by factors such as illumination,camera angle,photo distance and the size of the recognized object under natural conditions,it is difficult to guarantee the recognition accuracy and speed.A visual vocabulary tree retrieval method based on AKAZE fused with spatial location information is proposed.The AKAZE algorithm is used to extract image local features,quantify image local features into visual words through hierarchical word trees,and convert image retrieval into visual word statistical matching,the quantification process incorporates the spatial location information of local features and refines the importance of matching between features.The experimental results show that the method can improve the speed and accuracy of natural feature recognition,and can be further applied to indoor and outdoor as well as more complex interactive environments.(2)Due to the traditional network tracking and registration technology based on natural features,it is necessary to extract and match from each video frame,which not only requires a large amount of computation,but also is easily affected by the surrounding environment.In this paper,3D tracking combined with AKAZE and cone optical flow method is adopted.The AKAZE method is used to extract the feature points of the reference and video frames in the video stream,and then the features of the reference image and video frames are matched and screened for high-quality matching results,and then the tracking is performed by the pyramid optical flow method,and the motion estimation is performed to calculate the transformation matrix results,and finally the three-dimensional tracking registration is realized.The experiments show that the method can achieve accurate 3D registration and virtual-real fusion when there are viewpoint changes,rotation changes and occlusion situations,and can effectively reduce the impact of complex scenes on target tracking.(3)Based on the significance and current situation of WebAR applications in museums,art galleries and libraries,a WebAR system for libraries and art galleries is constructed based on the above proposed technology.It is verified that this system has good application value.