摘要
结合Gabor小波、二维线性鉴别分析(2DLDA)的特点,提出一种人脸识别方法。算法首先对人脸图像进行Gabor小波变换,然后进行2DLDA处理,最后使用最近邻法则进行分类。使用这种方法在ORL、Yale人脸库上进行测试,结果表明,Gabor-2DLDA方法比其它传统方法具有更优的性能,而且在提高识别率的同时算法的复杂程度并没有明显增加。
Combined with Gabor wavelet and Two-Dimensional LDA （2DLI）A）,a novel method for face recognition is presented. First,Gabor wavelet is used to transform the face images.Then 2DLDA is used to compress the feature.Finally,the nearest neighbor rule is used to classify,ORL and Yale face databases are used to test this new method.h shows that the recognition rate is higher than any other conditional method and the computational complexity did not increase significantly.
出处
《计算机工程与应用》
CSCD
北大核心
2008年第35期179-181,共3页
Computer Engineering and Applications
