基于神经网络的旗语识别系统

视觉是人类从外界获取信息的主要途径，人们希望计算机视觉在人机交互技术中发挥重要的作用。介绍一个基于视觉的动作实时识别系统，该系统用于在门桥模拟训练中对旗语动作进行识别。与以往工作不同的是，该系统仅用一个摄像机来实现多个动作的识别与跟踪，避免了多目视觉的特征匹配问题，进而提高了系统的稳定性和识别速度。

Flag Language Recognition System Based on ANN

Vision is a main approach to human being to get information. So people want computer vision to play an important role in human computer interaction. This paper describes a real-time vision based system for pontoon signer's training through gesture recognition and tracking. Stable detection can be achieved by extracting two kinds of features: statistic based feature and contour-based feature. Unlike most of previous works, our system recognizes hand gesture with just one camera, therefore avoiding the problem of matching image features between different views. So the stability and speed of recognition are improved greatly.