融合DWT与2DLDA的人耳识别  

Ear recognition combined DWT with 2DLDA

作　　者：屈智成 阳梦怡 魏玲江 

机构地区：[1]四川农业大学旅游学院,四川都江堰611830

出　　处：《华中师范大学学报（自然科学版）》2013年第6期786-789,共4页Journal of Central China Normal University：Natural Sciences

摘　　要：根据离散小波变换(DWT)和二维线性鉴别分析(2DLDA)的特点,结合人耳图像特征,提出一种融合DWT与2DLDA的人耳图像识别方法.该方法首先对人耳图像进行DWT,选择其中的低频子带,然后再在低频子带中利用2DLDA提取人耳图像特征,最后使用最近邻法则进行分类.实验结果表明,该方法识别效果优于2DLDA方法.According to the characteristic of discrete wavelet transform （DWT） and two dimensional linear discrimination analysis （2DLDA）,combined with human ear image features,a recognition method of the human ear image is put forward,which combines the DWT and 2DLDA.First of the method,decompose the human ear image using DWT to choose the low frequency subband,then extract the features of human ear image using 2DLDA in the low frequency subband,finally use the nearest neighbor rule to classify the human ear image.The experimental results show that the recognition effect of this method is better than 2DLDA.

关 键 词：离散小波变换 二维线性鉴别分析 人耳识别 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]