摘要
针对提取的指静脉感兴趣区域图像特征不明显的问题,提出一种基于多方向卷积模板和卷积神经网络的指静脉识别算法,采用多方向卷积模板对预处理后的指静脉图像进行骨架提取获取初步特征图,再通过轻便型卷积神经网络进一步提取特征,最后通过softmax分类器对样本进行训练.实验结果表明,该算法比现有的多数算法更准确.
Aiming at the problem that the image features of the region of interest of finger vein are not obvious,a finger vein recognition algorithm based on multi-directional convolution template and convolutional neural network was proposed.A multidirectional convolution template was used to extract the skeleton of the preprocessed finger vein image and obtain the preliminary feature map.Then,feature extraction was further carried out through the portable CNN network.Finally,the samples were trained by using softmax classifier.The experimental results show that the proposed method has higher accuracy than the existing related methods.
作者
吴叶清
WU Yeqing(Department of Information Engineering,Chenyi University College,Jimei University,Xiamen,Fujian 361021,China)
出处
《宜宾学院学报》
2020年第6期46-49,58,共5页
Journal of Yibin University
基金
福建省教育厅中青年教师教育科研科技类项目(JT180883)。
关键词
指静脉识别
卷积神经网络
多方向卷积模板
finger vein recognition
convolutional neural network
multidirectional convolution template