SURF特征及预处理RANSAC算法在人脸识别中的应用

摘要/Abstract

摘要： 针对人脸识别问题,提出了一种基于SURF特征的人脸图像快速识别方法。首先,对经预处理后的人脸图像提取SURF特征点,采用最近邻匹配法对特征点进行粗匹配；其次,利用KMeans聚类算法对粗匹配的特征点进行预处理来过滤明显不合适的匹配点,再利用RANSAC算法对过滤后的特征点实现精匹配,以达到对人脸的特征点比较准确地识别匹配。实验结果表明,该方法适用于手机终端的人脸图像的快速匹配,具有较强的鲁棒性及一定的实用价值。

关键词: SURF,特征点,RANSAC,人脸匹配,预处理,聚类算法,KMeans

Abstract: In the face recognition problem,a fast recognition method based on SURF features was proposed.Firstly,the SURF feature points are extracted from the preprocessed face image,and the nearest neighbor matching method is used for coarse matching of feature points.Secondly,the feature points of the coarse matching are processed by KMeans clustering algorithm to filter out the apparently inappropriate matches.Then RANSAC algorithm is used to achieve the precise matching of filtered feature points,in order to achieve accurate matching of recognition of face feature points.The experimental results show that the proposed method is suitable for the fast matching of face images in the mobile phone terminal,and has strong robustness and practical value.

Key words: SURF,Feature point,RANSAC,Face recognition,Pretreatment,Clustering algorithm,KMeans

蒋凌志. SURF特征及预处理RANSAC算法在人脸识别中的应用[J]. 计算机科学, 2015, 42(Z11): 209-212. https://doi.org/

JIANG Ling-zhi. Application of SURF Feature and Preprocessing RANSAC Algorithm in Face Recognition[J]. Computer Science, 2015, 42(Z11): 209-212. https://doi.org/

参考文献

[1] Frstner W,Gülch E.A fast operator for detection and precise location of distinct points,corners and centres of circular features[C]∥Proceedings of the ISPRS Intercommission Workshop.Interlaken,1987:149-155
[2] Harris C,Stepbeos M J.A combined corner and edge detector[C]∥Proc of the 4th Alvey Vision Conference.1988:147-151
[3] Zhang Jie-yu,Chen Qiang,Liu Fu-chang,et al.An improved robust estimation method of M-Estimators fundamental matrix [J].Journal of Image and Graphics,2009,4(8):1663-1668
[4] Bay H,Tuyteplaars T,van Gool L.SURF:speededup robustfeatures [C]∥Proceedings of the European Conference on ComputerVersion(ECCV2006).2006:404-417
[5] Juan L,Gwun O.A comparison of SIFT,PCA-SIFT and SURF[J].International Journal of Image Processing,2009,3(4):143-152
[6] Fischler M A,Bolles R C.Random sample consensus:apara-digm for model fitting with applications to image analysis and automated cartography [J].Communications of the ACM,1981,4(6):381-395
[7] Capel D P.Image mosaicing and super-resolution[D].Oxford:University of Oxford,2001:47-78
[8] Song Wei-yan.RANSAC algorithm and its application in remote sensing image processing[D].Beijing:North China Electric Power University,2011:3-9
[9] Haralick R M.Pose Estimation from Corresponding Point Data[J].IEEE Transactions on Systems,Man,and Cybernetics,1989,19(6)1426-1445
[10] 高健,黄心汉,彭刚.基于Harris角点和高斯差分特征点提取算法[J].模式识别与人工智能,2008,21(2):171-176
[11] 夏杰,李奇安,李悦,等.基于改进SIFT算法的图像匹配方法研究[J].电子设计工程,2012,20(14):157-160
[12] 王民,刘伟光.基于改进SIFT特征的双目图像匹配算法 [J].计算机工程与应用,2013,49(02):203-207
[13] 葛盼盼,陈强.基于SURF特征提取的遥感图像自动配准[J].计算机系统应用,2014,3(3):16-24
[14] 曲天伟,安波,陈桂兰,等.改进的RANSAC算法在图像配准中的应用[J].计算机应用,2010,0(7):1849-1851,2
[15] 谢娟英,高红超.基于统计相关性与K-means的区分基因子集选择算法[J].软件学报,2014(9):2050-2075
[16] 杨帆,邓振生.直方图均衡化与SURF重构的图像特征提取方法[J].计算机工程与应用,2013,49(10):188-191
[17] 赵璐璐,耿国华,李康,等.基于SURF和快速近似最近邻搜索的图像匹配算法[J].计算机应用研究,2013,30(03):921-923
[18] 尧思远,王晓明,左帅,等.基于SURF的特征点快速匹配算法[J].激光与红外,2014(3):347-350
[19] 尹龙,尹东,张荣,等.一种扭曲粘连字符验证码识别方法[J].模式识别与人工智能,2014(3):235-241

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed