计算机应用 ›› 2010, Vol. 30 ›› Issue (12): 3311-3313.

• 虚拟现实与模式识别 • 上一篇    下一篇

手形特征点定位方法

苑玮琦1,李燕2   

  1. 1.
    2. 沈阳工业大学
  • 收稿日期:2010-05-21 修回日期:2010-07-24 发布日期:2010-12-22 出版日期:2010-12-01
  • 通讯作者: 李燕
  • 基金资助:
    国家自然科学基金资助项目

Hand shape features location method

  • Received:2010-05-21 Revised:2010-07-24 Online:2010-12-22 Published:2010-12-01
  • Contact: LI Yan

摘要: 针对特征点定位的准确性影响识别精度,提出手形特征点的定位方法。首先采用基于直线拟合的方法确定手指的指尖点,然后基于方向跟踪的方法定位手指指根点,最后细定位指根点。基于特征矢量匹配算法的识别结果表明,人工测量识别率为84.62%,自动定位的识别率为84.35%。该方法能准确提取手形特征,具有可行性。

关键词: 特征定位, 直线拟合, 方向跟踪, 特征矢量, 手形识别

Abstract: The precision of hand shape identification is influenced by the accuracy of features location. Location method of feature point to carry out recognition based on relative finger length and width was presented. In this paper, first the fingertip was located based on linear fitting; then finger root was located by direction tracking; last the root points were determined. Experiments were carried out on HKUST, using the feature vector matching algorithm. Identification rate of 84.35%, the identification rate of actual measurement can reach 84.62%, and the effect of the proposed method is close to the manual measurement. The experimental results verify the validity of the proposed approach in personal authentication.

Key words: feature location, linear fitting, direction tracking, characteristic vector, hand shape recognition