计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (2): 164-167.

• 图形、图像、模式识别 • 上一篇    下一篇

四元数表示和NSCT变换的彩色人脸图像识别

李丙春,邓 勇   

  1. 喀什师范学院 网络中心,新疆 喀什 844000
    喀什师范学院 数学系,新疆 喀什 844000
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2012-01-11 发布日期:2012-01-11

Recognition of color face images based on quaternion matrices and NSCT transform

LI Bingchun, DENG Yong   

  1. Network Centre, KaShgar Teachers College, KaShgar, Xinjiang 844000, China
    Department of Mathematics, KaShgar Teachers College, KaShgar, Xinjiang 844000, China
  • Received:1900-01-01 Revised:1900-01-01 Online:2012-01-11 Published:2012-01-11

摘要: 将彩色人脸图像用四元数实矩阵的形式表示,对该矩阵作多级非抽样Contourlet(Nonsubsampled Contourlet,NSCT)变换,在各子带系数上作奇异值分解,进而获得图像的代数特征向量,由神经网络训练和识别。实验结果表明,该方法可以有效地提高彩色人脸图像的识别率,是一种有效的彩色人脸识别方法。

关键词: 四元数, 多级非抽样Contourlet变换, 奇异值, 神经网络, 彩色人脸识别

Abstract: In this paper, color face image is represented in quaternion real matrices. The coefficients in different scales and directions are obtained by image decomposition using the nonsubsampled contourlet transform. Singular value vector compressed is extracted as the color face image feature, and then the back propagation neural network is used to training and recognition. The experiment indicates that the recognition rate of the proposed method in this paper is improved and it is an effective color face image recognition method.

Key words: quaternion, Nonsubsampled Contourlet(NSCT) transform, singular values, neural network, color face recognition