基于细节点柱形编码和深度卷积特征的指纹检索方法

作者:宋德华;封举富 刊名:模式识别与人工智能 上传者:吴青青

【摘要】指纹检索方法使用细节点柱形编码作为特征,充分考虑指纹细节点的局部结构特征,却忽略指纹的整体结构特征,限制指纹检索的准确率.基于此种问题,文中提出基于细节点柱形编码和深度卷积特征的指纹检索方法.使用深度卷积网络学习指纹的整体结构特征(深度卷积特征),并结合深度卷积特征和细节点柱形编码,提升指纹检索的准确率.在3个经典指纹检索数据库上通过实验分析深度卷积特征的特性.实验表明,文中方法有效提升指纹检索的准确率.

全文阅读

基于细节点柱形编码和深度卷积特征的 指纹检索方法 宋德华1 封举富1 摘 要 指纹检索方法使用细节点柱形编码作为特征,充分考虑指纹细节点的局部结构特征,却忽略指纹的整体 结构特征,限制指纹检索的准确率.基于此种问题,文中提出基于细节点柱形编码和深度卷积特征的指纹检索方 法.使用深度卷积网络学习指纹的整体结构特征(深度卷积特征),并结合深度卷积特征和细节点柱形编码,提升指 纹检索的准确率.在3个经典指纹检索数据库上通过实验分析深度卷积特征的特性.实验表明,文中方法有效提升 指纹检索的准确率. 关键词 指纹检索,深度卷积神经网络,细节点柱形编码,特征表示 引用格式 宋德华,封举富.基于细节点柱形编码和深度卷积特征的指纹检索方法.模式识别与人工智能,2017, 31(2):175-181. DOI 10.16451/j.cnki.issn1003-6059.201802009 中图法分类号 TP391 Fingerprint Indexing Based on Minutia Cylinder-Code and Deep Convolutional Feature SONGDehua1,FENGJufu1 ABSTRACT Inthetypicalfingerprintindexingmethodbasedonminutiacylinder-code(MCC)feature, theminutiaelocalstructureisadequatelytakenintoaccount. Sincetheglobalstructureoffingerprintis ignored,theaccuracyoffingerprintretrievalislimited. Therefore,deepconvolutionalneuralnetworkis employedtolearntheglobalfeature(deepconvolutionalfeature)offingerprint. Then,theMCCanddeep convolutionalfeaturearefusedtoimprovethefingerprintindexingaccuracy. Experimentsarecarriedout to compare the proposed method with other prominent approaches on three benchmark databases. Besides,the property of deep convolutional feature is analyzed. Experimental results show that the proposedmethodeffectivelyimprovestheaccuracyoffingerprintindexing. Key Words FingerprintIndexing,DeepConvolutionalNeuralNetwork,MinutiaCylinder-Code,Fea- tureRepresentation Citation SONG D H,FENG J F. Fingerprint Indexing Based on Minutia Cylinder-Code and Deep ConvolutionalFeature. PatternRecognitionandArtificialIntelligence,2017,31(2):

参考文献

引证文献

问答

我要提问