基于图像处理技术的人机交互系统设计与实现

Design and implementation of human-machine interaction system based on image processing technology

基于混合脑电信号及视觉信息的人机交互系统无法对复杂环境中瑕疵图像进行有效修复,人机交互性能差。设计基于图像处理技术的人机交互系统,系统分为片上功能模块和外部电路模块,片上功能模块中的图像处理模块对图像进行预处理、图像还原以及采集特征点坐标。系统采用8255A芯片作为人机界面的硬件接口。该芯片A,B,C口分别连接数据线、控制线和行列线,实现人机交互界面信息的传输。系统通过带有USB的摄像头将图像转换成数字信号发送给计算机后,通过维纳滤波算法对图像进行修复。实验结果表明,所设计系统能够准确完成应用者的动作指令,具有较高的指令识别正确率和识别速度,人机交互性能佳。

Since the human-machine interaction system based on mixed EEG signals and visual information cannot effec-tively repair the defective images in complex environments,and has poor human-machine interaction performance,a human-ma-chine interaction system based on the image processing technology is designed.The system is divided into the on-chip functional module and external circuit module.The image processing module in the on-chip functional module preprocesses images,re-stores images,and collects the feature point coordinates.In the system,the 8255A chip is used as the hardware interface of the human-machine interface.The A,B and C interfaces of the chip are respectively connected to the data line,control line,and alignment line,so as to realize information transmission of the human-machine interaction interface.The system converts images into digital signals by using a camera with a USB.After sending the digital signals to the computer,images are repaired by using the Wiener filtering algorithm.The experimental results show that the designed system can complete the applicator′s action in-structions accurately,has a high instruction recognition accuracy rate and recognition speed,and has good human-machine inter-action performance.