Research On The Application Of Augmented Reality Technology In Intelligent Operation And Maintenance System Of Communication Networ

Augmented reality is a cutting-edge computer vision application technology in the intelligent age,which has been gradually applied to various fields,among which the communication network operation and maintenance system is an important research direction.3D registration and virtual-real fusion are the key technologies to realize augmented reality.3D registration is to estimate the camera pose in real time and obtain the position of the virtual object relative to the real scene according to 3D-2D point pairs.Virtual and real fusion is to match the corresponding virtual objects according to the 3D registration information,and realize the fusion of virtual information and real scene.Existing augmented reality technologies are faced with such problems as large delay in image recognition,large error in pose estimation,and slow human-computer interaction response when applied to the communication network operation and maintenance system.This paper focuses on how to solve the problem of image recognition,pose estimation and mobile terminal presentation technology when augmented reality technology is integrated into the actual operation and maintenance system.The main research work of this paper is as follows:(1)An improved SIFT(Scale Invariant Feature Transform)algorithm based on the equally spaced ring descriptor is proposed.In order to solve the problems of low precision and data redundancy in traditional SIFT algorithm,a circular descriptor composed of three equally spaced circles is used to replace the rectangular descriptor,so as to improve the rotation invariance of the algorithm.The precision of the algorithm can be improved by setting different weights to the ring according to the distance between the reference point and the key point.The improved SIFT algorithm assigns three thresholds to the three ring regions,and matches the best thresholds by constructing the mathematical model of the algorithm.Experimental analysis shows that,compared with the traditional SIFT algorithm,the improved algorithm has a great improvement in accuracy,calculation time efficiency,matching accuracy and rotation invariance.(2)A poth-DLS(Pose from Orthography and Scaling with Iterations-Direct Least Squares)camera position and Pose estimation technology based on expectation convergence is proposed.To solve the problems of pose estimation techniques for iterative method for a long time and low precision of iterative method,optimize the traditional POSIT algorithm.The improved algorithm each iteration of the input is determined by the expectations of part of the iterative output before,construct the hash table stores part of the input and output key/value pairs,use bloom filter to optimize hash table.Thus,the accuracy and time efficiency of the algorithm are improved.After the initial value of the camera pose was calculated by the direct least square method,the improved Posit algorithm was used to estimate the camera pose quickly and accurately.The experimental test results show that,compared with most of the Posit and other iterative methods,the improved algorithm significantly improves the computational time efficiency,while compared with most of the non-iterative methods,the algorithm accuracy is significantly improved.(3)Design and implement the intelligent operation and maintenance function of the communication network based on augmented reality.Using the improved SIFT image recognition algorithm and DLS-Posit camera position and pose estimation technology proposed in this paper,the following functions are designed and realized in the actual communication network operation and maintenance scene: Constructed 3D virtual simulation models,Developed mobile terminal software for intelligent operation and maintenance of communication network,realized virtual-real fusion,human-computer interaction,three-dimensional registration and other augmented reality operations with the handheld device system as the operating platform;Developed interactive Web pages based on Spring Boot,and realized remote monitoring,virtual information display and background support on the Web side.The software test results show that the AR application program designed in this paper based on two different architectures of mobile and Web terminal has normal function,high compatibility and robustness,and good user experience.