论文部分内容阅读
为提高半虚拟现实座舱中用户视觉真实感与人-座舱交互性能,提出了一种增强半虚拟现实座舱技术,将手图像实时融合到虚拟座舱环境中,使用户能实时看到自己的真实手。通过肤色检测算法实时分割摄像机采集图像,应用改进的自适应加权立体匹配算法与视图变形算法对图像进行视点校正,通过虚实匹配与虚实融合将手图像无缝融合到虚拟环境中,为用户提供增强虚拟环境中的可视化图像。测试结果表明:增强半虚拟现实座舱技术方案中触觉与视觉反馈真实,在头盔式虚拟环境中用户交互能点击到的最小方块边长为8mm。采用增强半虚拟现实技术方案完成座舱内典型交互动作的时间比传统数据手套方案与半虚拟交互方案平均分别降低约50%与18%,极大提高了人-座舱交互性能,可有效增强虚拟现实座舱系统实用性,并降低开发成本。
In order to improve the user visual reality and the human-cockpit interactive performance in the semi-virtual reality cockpit, an enhanced semi-virtual reality cockpit technology is proposed, which realizes the integration of hand images into the virtual cockpit environment in real time so that the user can see his real hand in real time . The real-time segmentation camera is used to acquire the image through the skin color detection algorithm. The improved adaptive weighted stereo matching algorithm and the view deformation algorithm are used to correct the viewpoint of the image. The hand image is seamlessly fused to the virtual environment through the virtual-reality matching and the virtual-reality fusion, Visual image in a virtual environment. The test results show that the real tactile and visual feedback in the solution of semi-virtual reality cockpit is real, and the minimum side length of the square that user interaction can click in the helmet-like virtual environment is 8mm. Compared with the traditional data glove scheme and semi-virtual interaction scheme, the average half-virtual reality solution reduces the average interaction time by about 50% and 18% respectively, which greatly improves the human-cockpit interactive performance and effectively enhances the virtual reality Cockpit system practicality, and reduce development costs.