Study on Augmented Reality Method for Unlabeled Recognition Based on Filtering Processing Optimization

2018 2nd annual International Conference on Cloud Technology and Communication Engineering

计算机科学;无线电电子学

Chen, Sihao^1

College of Information Science, Beijing Language and Culture University, Beijing

100083, China^1

关键词: Filtering processing;    Information amount;    Matching algorithm;    Recognition algorithm;    Recognition efficiency;    Two dimensional images;    Two dimensional plane;    Virtual information;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/466/1/012103/pdf
DOI  :  10.1088/1757-899X/466/1/012103

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

Augmented Reality (AR), as a technology that superimposes virtual objects and real scenes, expands the information amount of actual scenes through virtual information. In view of the shortcomings of the previous AR application methods, this paper proposes an AR application implementation method of unlabeled recognition based on filtering processing optimization. This algorithm uses filtering to preprocess and optimize the image, and solves the problem of poor matching effect of the previous recognition algorithm. The 3D model is established in the two-dimensional image and the model is rendered on the two-dimensional plane. The experimental results show that compared with the previous matching algorithm, the recognition efficiency of this method is improved by more than 10%, which has more obvious advantages in matching efficiency and better robustness.