Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (14): 209-211.DOI: 10.3778/j.issn.1002-8331.2010.14.062

• 图形、图像、模式识别 • Previous Articles     Next Articles

Facial contour extraction based on GVF Snake model

SONG Shu-zhong,GAO Chun-geng,MA Jian-wei   

  1. Electronic Information Engineering College,Henan University of Science & Technology,Luoyang,Henan 471003,China
  • Received:2008-10-22 Revised:2009-01-17 Online:2010-05-11 Published:2010-05-11
  • Contact: SONG Shu-zhong

基于GVF Snake模型的人脸轮廓提取

宋书中,高春庚,马建伟   

  1. 河南科技大学 电子信息工程学院,河南 洛阳 471003
  • 通讯作者: 宋书中

Abstract: Active contour model is an important research domain in computer vision.Aiming at the limitations that the traditional Snake model cannot converge to concave edge and the original contour must be close to the real edge of the image,a new facial contour extraction method based on Gradient Vector Flow(GVF) Snake model is proposed by improving the external energy.The algorithm uses GVF as its external energy.It not only expands the effective range of the energy field but also enhances the ability to deal with the concave edge.Experiments show that it can extract facial contour fast and exactly.

Key words: facial contour extraction, Snake model, gradient vector flow

摘要: 活动轮廓模型是计算机视觉领域的重要研究方向。针对传统的活动轮廓模型(Snake模型)对凹形轮廓处理效果差、初始轮廓必须充分接近图像边缘的缺点,通过改进外部能量项,提出了一种基于梯度矢量流活动轮廓模型的人脸轮廓提取算法。该算法把梯度矢量场作为外部能量场,克服了传统Snake模型力场范围小以及不能收敛于凹形边缘的缺点。实验结果表明,该方法能够快速、准确地提取人脸轮廓。

关键词: 人脸轮廓提取, Snake模型, 梯度矢量流

CLC Number: