一种基于深度计算的图像增强现实系统实现  

Implementation of an Image Augmented Reality System Based on Depth Calculation

作　　者：翟丽杰[1] ZHAI Lijie(Department of Physics and Electrical Engineering,Weinan Normal University,Weinan 714099,Shaanxi China)

机构地区：[1]渭南师范学院物理与电气工程学院,陕西渭南714099

出　　处：《河南科学》2020年第2期188-193,共6页Henan Science

基　　金：2019年陕西省教育厅专项科研项目(19JK0288)。

摘　　要：为解决在低能见度条件下由于景物无法辨识而无法进行增强现实的问题,以及在增强现实中虚拟物体与真实物体之间的深度一致性问题,研究了增强现实技术原理,给出了低能见度条件下增强现实方案,并设计了一个基于深度计算的图像增强现实系统.该系统可利用双目摄像机实时采集目标图像,并利用校正标定后得到的视差图像,采用半局部匹配方法(SGBM)算法和深度计算实现虚拟物体的跟踪配准.在不同环境下的实验表明,该系统具有良好的实时性、光不变性和深度一致性.In order to solve the problem that the scene can not be identified in low visibility and the depth consistency between the virtual object and the real object in augmented reality,the principle of augmented reality technology is studied. The augmented reality scheme in low visibility is given,and image augmented reality system based on depth calculation is designed. The system can use binocular camera to collect real-time target image,and get better parallax image after calibration. The Semi Local Matching Method(SGBM)algorithm and depth calculation are used to realize the tracking and registration of virtual object. Experiments in different environments show that the system has good real-time,light invariance and depth consistency.

关 键 词：低能见度 增强现实 深度计算 深度一致性 半局部匹配方法 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术] TP317[自动化与计算机技术—计算机科学与技术]