基于图像分块的LDA人脸识别  

LDA for face recognition based on image segmentation

作　　者：王文豪[1] 严云洋[1] 

机构地区：[1]淮阴工学院,江苏淮安223001

出　　处：《计算机工程与设计》2007年第12期2889-2891,共3页Computer Engineering and Design

摘　　要：设计了一种基于图像分块的LDA(linear discriminant analysis)人脸识别方法,该方法从模式的原始数字图像出发,先对图像矩阵进行分块,然后对分块子图像进行LDA特征提取,从而得到能代替原始模式的低维新模式,最后再用最小距离分类器进行分类。该方法克服了传统LDA方法的缺点,其优点是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Fisherfaces方法。A method of LDA for face recognition is presented, based on image segmentation. Firstly, in the method, the original images are divided into block images, which are also called sub-images. Then, the well-known LDA method is directly employed to the sub-images obtained from the previous step, and the new lower dimensionality patterns that replace the original patterns are obtained. Finally the recognition results are obtained by the general minimum distance classifier. This method is introduced to overcome the drawbacks existing in the traditional LDA method. The advantage of the represented way is that the local discriminant features of the original image is efficiently extracted. The experimental results show that the proposed method is superior to Fisherfaces at the recognition rates.

关 键 词：人脸识别 图像分块 线性鉴别分析 小样本问题 最小距离分类器 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]