摘要
针对手指静脉特征提取及匹配识别问题,设计了一套嵌入式小型化手指静脉采集装置,并提出了一种基于散射卷积网络算法的手指静脉识别方法.对采集到的原始手指静脉图像进行感兴趣区域提取和预处理,利用多层散射卷积网络提取每张图像的散射能量分布特征,计算每个子块图像能量均值和方差作为特征向量,利用支持向量机进行样本训练和匹配识别.实验结果表明:该方法用于手指静脉识别相比于目前的其他方法能有更好的效果,识别率达到100%.
In order to extract and recognize the feature of finger vein,this paper designs a compact embedded device for acquiring finger vein images,and proposes a method of finger vein recognition based on scattering convolution network algorithm. Firstly,the interested regions in the original finger vein images are extracted and preprocessed. Secondly,multi-layer scattering convolution network is used to extract the scattering energy distribution feature matrix for each image. Each sub-block image energy mean and variance is calculated as the feature vector. Finally,the SVM is used to train the samples and match recognition. The experimental results show that the method used in the finger vein recognition is better than the other methods and accuracy rate of the algorithm reach 100%.
作者
陈朋
姜立
王海霞
陈培
CHEN Peng;JIANG L i;WANG Haixia;CHEN Pei(College of Information. Engineerings Zhejiang University of Technology？ Hangzhou 310023？ China)
出处
《浙江工业大学学报》
CAS
北大核心
2018年第1期56-60,共5页
Journal of Zhejiang University of Technology
基金
国家自然科学基金资助项目(61527808,61602414)
关键词
生物识别
手指静脉
散射卷积网络
散射能量
支持向量机
biological recognition
finger vein
scattering convolution network
scattering energy
support vector machine
