基于高通滤波及LBP主成分特征的一种掌纹识别方法

本文用高斯高通滤波对图像进行预处理，然后用LBP方法提取掌纹图像特征，最后用PCA法降低特征维数。高斯高通滤波的作用在于增强图像对比度，使其具有更为明显的区分信息；图像的LBP特征具有抗旋转能力强，不受每次采集图像时光照不同的影响等优点；PCA能够提取特征矩阵的主成分。试验证明此方法具有较好的特征提取能力，得到了较高的识别率。

Palmprint Recognition Based On Highpass Filter and Principle Components' Character of LBP

In this paper, we use gauss highpass filter to pre-process the image, then use LBP to extract the palmprint images＇ features, and use PCA to reduce the dimension of features in the last. The pre-process play a role in enhance the image, which made it has more distinctive messages; the LBP character has a strong power of anti-rotation, and also has the ability to avoid of the images＇ different light that took in different time; the PCA could abstract the principle component of the matrix. This method has a better ability to abstract the features, and get a higher recognized rate.