摘要
提出了一种基于图像分块的二维保局投影(分块2DLPP)的人脸识别方法。先对原始图像矩阵进行分块,然后对分块子图像施行2DLPP方法,再将各个分块按照一定的次序整合起来进行特征提取,从而实现图像降维。该方法能有效地提取图像的局部特征。实验表明:该方法在识别性能上优于2DLPP方法。
A method of Two-Dimensional Locality Preserving Projections (2DLPP) for face recognition was proposed, based on modular image. Firstly, the original images were divided into modular images by presented approach. Secondly, the 2DLPP method was applied to the sub-images obtained from the previous step. Then, the modular images were combined according to a certain order to extract the features. Therefore, the dimension of the original images could be depressed. This approach could distill the local features of the images effectively. The experimental results indicate that the proposed method is superior to 2DLPP in recognition performance.
出处
《计算机应用》
CSCD
北大核心
2009年第8期2056-2059,共4页
journal of Computer Applications
关键词
人脸识别
图像分块
特征提取
二维保局投影
face recognition
modular image
feature extraction
Two-Dimensional Locality Preserving Projections (2DLPP)