In the human–computer interaction field, a contactless interaction with large screens through gestures is very representative, and the recognition and filtering of gesture images are very important tasks. Aiming at solving the problems of interference and positioning drift of three-dimensional lidar sensors, this article proposes a contactless interactive control system based on switching filtering algorithm, which selects the Butterworth filtering and the modified strong tracking Kalman filter to be used in the filtering process. The proposed interactive system extracts and optimizes user gestures, maps the gestures to the screen, simulates mouse operations, and enables operations such as selection, sliding, zooming in and out, and others. This switching filtering algorithm effectively solves the accuracy problem of a single filtering algorithm and the rapidity of complex filtering algorithms in the signal processing step, and greatly improves the interaction accuracy without sacrificing too much processing time. The experimental results show that by applying the proposed switching filtering algorithm to a contactless human–computer interaction system, the system can achieve smooth gesture interaction. The proposed system can perform real-time interaction with multiple people, which fully verifies the effectiveness and superiority of the proposed algorithm.

在人机交互领域，通过手势与大屏幕进行非接触式交互非常具有代表性，并且手势图像的识别和过滤是非常重要的任务。为了解决三维激光雷达传感器的干扰和定位漂移问题，提出了一种基于开关滤波算法的非接触式交互式控制系统，该系统在滤波过程中选择了Butterworth滤波和改进的强跟踪卡尔曼滤波。 。所提出的交互式系统提取并优化用户手势，将手势映射到屏幕，模拟鼠标操作，并启用诸如选择，滑动，放大和缩小等操作。这种切换滤波算法有效地解决了单一滤波算法的精度问题和复杂滤波算法在信号处理步骤中的快速性问题，并且在不牺牲过多处理时间的情况下，极大地提高了交互精度。实验结果表明，将所提出的开关滤波算法应用于非接触式人机交互系统中，该系统可以实现流畅的手势交互。所提出的系统可以与多个人进行实时交互，从而充分验证了所提出算法的有效性和优越性。实验结果表明，将所提出的开关滤波算法应用于非接触式人机交互系统中，该系统可以实现流畅的手势交互。所提出的系统可以与多个人进行实时交互，从而充分验证了所提出算法的有效性和优越性。实验结果表明，将所提出的开关滤波算法应用于非接触式人机交互系统中，该系统可以实现流畅的手势交互。所提出的系统可以与多个人进行实时交互，从而充分验证了所提出算法的有效性和优越性。