摘要: 提出了一种新的图像距离—— 图像匹配距离(Image Matching Distance,IMMD).IMMD考虑图像中每个像素与对应图像特定区域的关系,在特定区域寻找与该像素匹配的点,从而将图像的灰度值及其坐标位置引入到图像的相似性度量中.这样使得IMMD对人脸姿态、表情、角度变化具有较好的鲁棒性.用基于图像匹配距离的最近邻分类器进行人脸识别.实验结果表明,基于IMMD的方法优于基于传统欧氏距离和图像欧氏距离的同类型方法.
[1]Zhao W,Chellappa R,Rosenfel A,et a1.Face recognItion:a literature survey[J].ACM Computing,2003,35(4):399-458.[2]李乐,章毓晋.利用符号假设检验比较两类PCA人脸识别算法[J].模式识别与人工智能,2008,21(4):433-440.[3]Haitao Zhao.Incremental linear discriminant analysis for face recognition[J].IEEE Transactions on systems,man,and cybernetics.Part B,Cybernetics,2008,38(1):210-221.[4]Yang J,Zhang D,Frangi A F,et a1.Two—dimensional PCA:a new approach to appearance based face repr-esentation and recognition[J].IEEE Trans PaRem An—alysis and Machine InteHigence,2004,26(1):131-1 37.[5]Keun—Chang Kwak.Face recogmtion using an enhanced independent component analysis approach[J].IEEE Transactions on Neural Networks,2007,18:530-541.[6]Li Wei,Yan Zhang,Feng Ju—fu.On the euclidean distarice of images[J].IEEE Transactions On Pattern Anal-ysis and Maching Intelligence,2005,27(8):1334-1339.[7]刘福新,杜世培,陈益强.基于改进Hausdorf距离的人脸匹配方法[J].计算机工程与应用,2007,43(35):169-171.[8 j Daniel Keysers.Adaptation in statistical pattern recognition using tangent vectors[J].IEEE Transa-ctions on Pattern Analysis And Machine Intelligence,2004,26(2):269-274. |
No related articles found! |
|