一种自动的人脸轮廓定位方法*

南京大学学报(自然科学版) ›› 2013, Vol. 49 ›› Issue (2): 183–188.

一种自动的人脸轮廓定位方法*

李听听¹，龚勋²**，夏冉³

出版日期:2015-10-30 发布日期:2015-10-30
作者简介: (1.四川大学锦城学院计算机科学与软件工程系，成都,611731;2.西南交通大学信息科学
与技术学院，成都,610031;3.重庆邮电大学计算机科学与技术研究所，重庆，400065)
基金资助:
国家自然科学基金(61202191)，中央高校基本科研业务费专项资金(SWJTU12CX095)

An automatic face contour extracting method

Li Xin一Xin¹，Gong Xun²，Xia Ran³

Online:2015-10-30 Published:2015-10-30
About author: (1 .Department of Computer Science and Software Enginecring,Jincheng College of Sichuan University,Chengdu
611731,China;2. School of Information Science and Technology, Southwest Jiaotong University,
Chcngdu，610031,China;3.lnstitute of Computer Science and Technology,Chongqing
University of Posts and Telecommunications,Chongqing,400065,China)

摘要/Abstract

摘要： 人脸分割对人脸识别、人脸三维建模等人脸图像处理问题具有重要意义，而人脸图像往往轮
廓边缘模糊、梯度不明显，常规无边缘几何活动轮廓模型通常无法获得理想的分割效果且计算量较大.
为实现快速、准确的人脸轮廓定位及分割，将无边缘几何活动轮廓模型和稀疏场数值算法相结合提出了
一个改进的算法，并结合人脸检测和数学形态学算子提出一个基于曲线演化的人脸分割方案.实验结果
表明，该算法不仅提高了计算效率，还可以有效地检测出局部模糊或分断边界，进化曲线不会断裂，能够
获得较好的人脸分割效果.

Abstract: Images containing faces are essential to intelligent vision-based human computer interaction,and research
efforts in face processing include face recognition, face tracking, and expression recognition. Many applications as
sume that the faces in an image or an image sequence have been identified and localized.To build fully automated sys
tans that analyze the information contained in face images，robust and efficient face detection algorithms are re-
quired. However, such a problem is challenging because faces are norrrigid and have a high degree of variability in
sizc，shape，color,and texture.The purpose of this paper is to provide a relative robust method for face segmentation
in images based on curve evolution methodology. Since the face image always has a blur boundary and little gradient
changes,thc region segmentations obtained by the original Chan-Vese model arc generally unsatisfactory and need
large amount of calculations.To achieve more accurate facial contour extraction and face segmentation, a new face
segmentation scheme based on curve evolution model is proposed which is a combination of Chan-Vese model，spars
field algorithm, face detection and mathematical morphology operators. Experimental results show that the improved
algorithm can effectively detect the local blur and breaking boundaries on the face images without any fractures in the
curve,hence resulting in favorable face segmentations.

李听听¹，龚勋²**，夏冉³. 一种自动的人脸轮廓定位方法*[J]. 南京大学学报(自然科学版), 2013, 49(2): 183–188.

Li Xin一Xin¹，Gong Xun²，Xia Ran³
. An automatic face contour extracting method[J]. Journal of Nanjing University(Natural Sciences), 2013, 49(2): 183–188.

参考文献

[1]Wu L F, Shen L S, Kong X , et al. Face segmentation based on curve fitting. Chinese Journal of Computers,
2003,26(7);893-897.(毋立芳，沈兰荪，孔潇等. 基于曲线拟合的人脸分割算法.计算机学报，2003, 26(7) :893一899).
[2]Govindaraju V. Locating human faces in photographs.International Journal of Computer Vision, 1996，19 :129一146.
[3]Chen D S, Liu Z K. Edge features in color image and their face detection performance evaluation. Journal of Software, 2005，16(5):727一732.(陈锻生，刘政凯.彩色图像边缘特征及其人脸检测性能评价.软件学报，2005,16(5);727-732).
[4]Osher S, Sethian J A. Fronts propagating with curvature dependent speed:Algorithms based on
HamiltorrJacobi formulations. Journal of Compu- tational Physics, 1988，79(1):12一49.
[5]Sethian J A. Level set methods and fast marching methods;Evolving interfaces in computational ge ometry,fluid mechanics，computer vision,and ma-terials science. Cambridge, United Kingdom: Cambridge University Press，1999，378.
[6]Kass M, Witkin A, Zcrzopoulous D. Snakes; Active contour models. Proceedings of the First international
Conference on Computer Vision. London;lEEE Corm puter Society Press,1987;259一268.
[7]Caselles V, Kimmel R, Sapiro G. Gcodesic active contours. International Journal of Computer Vision, 1997,22(1):61一79.
[8]Chan T F, Vese L A.Active contours without ed- ges. 1EEE Transactions on Image Processing, 2001, 10(2):266一277.
[9]Happer P, Reilly R B. Color based video segmenta- tion using level sets. Prececdings of the international
Conference on image. Vancouver;Print’n Run Press, 2000,480一483.
[10]Li S J , Wang W U, Xic P, et al. An adaptive level set model with feature selection for remote sensing image
segmentation. Journal of Nanjing University ( Natural Sciences),2012,48(4):482 -490.(李士进，王万国，
谢萍，万定生等.面向遥感图像分割的自适应特征选择水平集模型.南京大学学报(自然科学)，2012,48(4) : 482一490).
[11]Huang F Z,Shu J B,Xi Y G. Geometric active contour for face contour extraction. Journal of Image and Graphics,2003,5(8):546一550.(黄福珍，苏剑波，席裕庚.基于几何活动轮廓模型的人脸轮廓提取方法.中国图象图形学报，2003,5(8):546 -550).
[12]Liu T，Zhang D L. A novel fast level set initializa- tion method. Journal of image and Graphics，
2010,15(5):775-781.(刘涛，张大龙.一种新颖的快速水平集初始化方法.中国图象图形学报，2010,15(5):775一781).
[13]Adalstein D,Sethian J. A fast level set method for propagating interfaces. Journal of Computational
Physic，1994，118(2);269一277.
[14]Whitaker R, Breen D, Museth K. A level-set approath to 3D reconstruction from range data. Journal of Corm
puter Vision,1998,29(3):203一231.
[15]Museth K，Breen D, Whitakcr R，et al. A Level set surface editing operators. ACM Transactions
on Graphics,2002,21(3):330一338.
[16]BioID L. BioID人脸数据库.http;//www. bioid. Com/technology/facedatabasc. htm1.2010-08-07.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed