一种改进的基于主动形状模型的人脸特征定位方法A Improved Facial Feature Localization Method Based On Active Shape Model
田建华
摘要(Abstract):
主动形状模型是目前一种常用的人脸特征定位方法.针对传统的主动形状模型过度依赖模型初始参数的设置问题,提出了一种改进的基于主动形状模型的人脸特征定位方法.首先,通过样本学习得到输入新图像的灰度重构系数,并将这组系数用于人脸形状的重构,再由重构出的人脸形状得到主动形状模型的初始参数,然后通过不断调整模型参数减少模型与目标轮廓的距离误差,最后在数次迭代后达到模型与实际人脸特征轮廓的匹配.与基于传统主动形状模型的特征定位相比,改进的主动形状模型具有较高的准确性,能快速定位出各目标特征.
关键词(KeyWords): 人脸特征定位;样本学习;主动形状模型
基金项目(Foundation):
作者(Author): 田建华
DOI: 10.16393/j.cnki.37-1436/z.2011.02.023
参考文献(References):
- [1]Cootes T,Taylor C,Cooper D,et al.Active shape models-Their training and application[J].Computer Vision and ImageUnderstanding,1995,61(1):38-59.
- [2]Cootes T F,Taylor C J.Active shape models—Smart snakes[C].Berlin:British Machine Vision Conference,1992:266-275.
- [3]Kass M,Witdin A,Terzopoulous D.Snakes:active contour Model[J].International Journal of Computer Vision,1987,1(4):321-331.
- [4]柴秀娟,山世光,高文.基于样例学习的面部特征自动标定算法[J].软件学报,2005,16(5):718-725.
- [5]Cootes T F,Hill A,Taylor C J,et al.The use of active shape models for locating structures in medical images[J].Image and vi-sion computing,1994,12(6):355-366.
- [6]Abdullah A.Al-Shaher,Edwin R.Hancork.Learning mixtures of point distribution models with the EM algorithm[J].PatternRecognition,2003,36(12):2805-2818.
- [7]Cootes T,Taylor C.Combining point distribution models with shape models based on finite element analysis[J].Image VisionCompute,1995,13(5):403-409.