融合视觉信息感知与虚拟现实的康复辅助系统

Rehabilitation Assistance System Based on Visual Information Perception and Virtual Reality

当前人工智能技术在康复领域大多用于严重运动障碍群体,而中国现阶段人口老龄化加剧,会导致轻型运动障碍人群的大量增加,亟需更加智能化的运动训练与康复系统.针对轻型运动损伤患者以及普通老年群体,本文提出了一种基于视觉的运动信号感知与虚拟现实的康复辅助系统,该系统通过实时推断人体骨骼模型,利用虚拟现实游戏与骨骼点运动进行交互,实现引导康复动作,最终对康复动作进行评价并给出交互界面提示.本文招募受试者参加实验,在闭环训练中受试者通过本文系统在虚拟康复游戏的引导反馈下复现标准康复动作,对比分析受试者实验结果,验证了所提出方法的有效性与可行性.

At present,artificial intelligence technology is mostly used in the field of rehabilitation for people with serious sports disabilities.However,the aging of the population in China at this stage will lead to a large increase in people with light sports disabilities.More intelligent sports training and rehabilitation systems are urgently needed.Aiming at the patients with light sports injuries and ordinary elderly groups,this paper proposes a rehabilitation assistance system based on visual perception of motion signals and virtual reality.This system uses virtual reality games to interact with skeleton point movements by inferring human skeleton models in real time to guide rehabilitation actions,and finally evaluates rehabilitation actions and gives interactive interface prompts.This paper recruited subjects to participate in the experiment.In the closed-loop training,the subjects reproduced the standard rehabilitation actions under the guidance and feedback of the virtual rehabilitation game through the system in this paper.The comparative analysis of the experimental results of the subjects verified the effectiveness and feasibility of the proposed method.