During the immersive virtual reality experience, because the visual senses are completely enclosed in the virtual environment, they are unable to perceive environmental changes in the real world and produce unsafe feelings, which affects their experience in virtual reality systems. In order to improve people's sense of security in the immersive virtual reality experience, the paper has designed four ways to interact with the real world in a virtual environment, so that users can get real‐world information in the virtual environment. This paper tests users' changes in their sense of security during immersive virtual reality experiences under the influence of various interaction methods and conduct psychological analysis. Twenty‐one volunteers are recruited to participate in the experiment, and their personal safety and psychological effects are tested. The experimental results show that the method of using Yolcat neural network to segment the captured images according to categories and then fuse the segmented images into the virtual environment can improve the user's sense of security without destroying the immersion.

中文翻译：

---

虚拟沉浸系统中融合显示模式对用户个人安全感的影响

在身临其境的虚拟现实体验中，由于视觉完全封闭在虚拟环境中，因此它们无法感知现实世界中的环境变化并产生不安全的感觉，从而影响他们在虚拟现实系统中的体验。为了提高人们在沉浸式虚拟现实体验中的安全感，本文设计了四种在虚拟环境中与现实世界进行交互的方式，以便用户可以在虚拟环境中获取现实世界的信息。本文在各种交互方法的影响下，在沉浸式虚拟现实体验中测试用户的安全感变化，并进行心理分析。招募了21名志愿者参加该实验，并对其人身安全和心理影响进行了测试。实验结果表明，利用Yolcat神经网络按类别对捕获的图像进行分割，然后将分割后的图像融合到虚拟环境中的方法，可以在不破坏沉浸感的前提下提高用户的安全感。