摘要
针对中国聋人手势词"语形"是由若干个基本手势组成的特点,本文提出了沿时间轴的贪心聚类算法,并在此基础上给出了一种快速训练算法及快速识别算法。将该算法具体应用到中国手语手势词的识别中,实验结果表明,与HMM相比,该方法不仅在识别速度上有大的改观,而且大大缩短了手势词对应模板的训练时间。
According to the characteristics of sign word of gesture language for Chinese Sign Language (CSL), a greedy clustering algorithm along the time axis (GLATA) is proposed in this paper. A fast training algorithm and a fast recognizing algorithm basing on GLATA are given, and they are applied in gesture recognition for CSL. The experimental results show that this method is more effective than HMM in gesture recognition.
出处
《模式识别与人工智能》
EI
CSCD
北大核心
2005年第1期1-5,共5页
Pattern Recognition and Artificial Intelligence
关键词
手势识别
沿时间轴贪心聚类算法
时间序列聚类
Gesture Recognition
A Greedy Clustering Algorithm along the Time Axis
Time Series Clustering