基于局部保持投影的掌纹识别

为了保持掌纹空间的局部结构,运用局部保持投影(LPP)方法进行掌纹识别.在小样本图像识别中,特征方程矩阵存在奇异性.传统的解决方法是运用主元分析(PCA)获得原样本的低维特征子空间,在该空间中运用LPP进行特征提取.由于PCA和LPP的投影标准本质上是不同的,PCA降维时丢失许多重要的判别信息.为了解决这个问题,提出运用三级小波变换、图像下抽样、图像分块求平均值三种方法降低掌纹空间的维数,在低维图像上应用LPP提取局部特征.计算特征矢量间的余弦距离进行掌纹匹配.运用PolyU掌纹图像库进行测试,结果表明,该算法的识别性能均优于PCA和PCA LPP.特征提取和匹配总时间小于0.1 S,具有快速、有效、易于实现等优点.

Palmprint Recognition Based on Locality Preserving Projection

In order to preserve the local structure of the image space,locality preserving projection(LPP) is applied to palmprint recognition.In small-size-sample cases such as image recognition,the matrix of the eigenvalue equation is singular.The traditional solution is to utilize the principal component analysis(PCA) as a pre-processing step aiming to reduce the dimensionality of the palmprint space,then LPP is applied to extract feature.Since the projection criterion of the PCA and that of LPP are essentially dif...