摘要
根据人体手指静脉图像的特点,提出了一种基于降维思想的手指静脉特征提取方法,即搜索二维灰度图像中的静脉在一维灰度曲线上形成的谷形区域来实现静脉特征的提取。为了对手指静脉特征的充分提取,在进行特征提取前,采用了高频强调滤波对图像进行增强处理。实验结果表明,该方法思想独特,简单易行,对手指静脉的特征提取达到满意的效果。
According to the features of human finger vein image, a finger vein feature extraction method based on dimensionality reduction idea is put forward. In this method, the vein feature extraction is carried out through searching for valley shape region that is formed by the vein of two dimension gray image on one dimension gray curve. In order to finish sufficient extraction of the finger vein feature, the image is enhanced by high frequency emphasize filter before feature extraction. The experimental results show that this method has particular idea. It is simple and easy to realize. The effect of finger vein feature extraction is satisfied.
出处
《计算机工程与设计》
CSCD
北大核心
2008年第16期4383-4384,4388,共3页
Computer Engineering and Design
基金
重庆市教育委员会科学技术研究基金项目(KJ070620)
关键词
生物识别
手指静脉
高频强调滤波
特征提取
图像识别
biology recognition
finger vein
high frequency emphasize filter
feature extraction
image recognition
