摘要
本文介绍了一种用于手势识别的新方法——图像的局部方向直方图矢量(OHV),利用图像的OHV作为手势的特征向量来进行手势的分类和识别。该方法对于光线和手势平移变化不敏感,具有较强的鲁棒性。在进行特征矢量匹配时,本文使用欧氏距离作为矢量间匹配程度的度量算法。实验结果表明该方法具有很高的识别率,并且简单,快速,可以用于实时的手势识别系统中。
In this paper, a new method for gesture recognition based on orientation histogram vector mapping is proposed. We use the orientation histogram as a feature vector for gesture classification and recognition. This method is simple and fast to compute, and offers some robustness to scene illumination changes and gesture translation. The experiments result shows that the recognition rate is high, we can applying this method to a real-time gesture recognition system.
出处
《信息技术与信息化》
2006年第2期53-55,共3页
Information Technology and Informatization
关键词
手势识别
方向直方图矢量
欧氏距离
Gesture Recognition
Orientation Histogram Vector
Euclidean distance metric
