一种自动的人脸轮廓定位方法  

An automatic face contour extracting method

作　　者：李昕昕[1] 龚勋[2] 夏冉[3] 

机构地区：[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

出　　处：《南京大学学报（自然科学版）》2013年第2期183-188,共6页Journal of Nanjing University（Natural Science）

基　　金：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)

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.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 assume that the faces in an image or an image sequence have been identified and localized.To build fully automated systems that analyze the information contained in face images,robust and efficient face detection algorithms are required.However,such a problem is challenging because faces are non-rigid and have a high degree of variability in size,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,the region segmentations obtained by the original Chan-Vese model are 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,sparse-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.

关 键 词：人脸分割 人脸轮廓提取 无边缘几何活动轮廓模型 水平集 稀疏场算法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]