基于改进卷积神经网络的指静脉识别
|
摘要
针对深度学习的指静脉识别算法在训练样本较少以及训练次数较低时算法识别率降低问题,提出基于改进卷积神经网络的指静脉识别算法。通过增加卷积层数并使用LeaKy ReLU作为激活函数,提高网络泛化能力;使用改进的池化模型降低网络特征维度;反向传播调整权值时,引入判别信息作为约束条件。实验结果表明,该算法准确率在训练样本较少以及训练次数较低时明显优于其它指静脉识别算法。
Aiming at the problem that the recognition rate of finger vein recognition algorithm for deep learning is low when training samples and training times are insufficient,the algorithm of finger vein recognition based on improved convolutional neural network was proposed.The number of convolutions was increased and LeaKy ReLU was used as the activation function to improve the generalization ability of the network.The improved pooling model was used to reduce the network feature dimension.When the weights were adjusted in reverse,the discriminant information was introduced as a constraint condition.Experimental results show that the accuracy of the proposed algorithm is obviously superior to other finger vein recognition algorithms when training samples and training times are insufficient.
Aiming at the problem that the recognition rate of finger vein recognition algorithm for deep learning is low when training samples and training times are insufficient,the algorithm of finger vein recognition based on improved convolutional neural network was proposed.The number of convolutions was increased and LeaKy ReLU was used as the activation function to improve the generalization ability of the network.The improved pooling model was used to reduce the network feature dimension.When the weights were adjusted in reverse,the discriminant information was introduced as a constraint condition.Experimental results show that the accuracy of the proposed algorithm is obviously superior to other finger vein recognition algorithms when training samples and training times are insufficient.
引文
[1]PANG Xiaohong.Research on the algorithm of finger vein identification[D]. Tianjin:Hebei University of Technology,2015(in Chinese).[庞晓红.指静脉身份识别算法的研究[D].天津:河北工业大学,2015.]
[2]CHEN Yaodan, WANG Lianming.Face recognition method based on convolution neural network[J].Journal of Northeast Normal University(Natural Science),2016,48(2):70-76(in Chinese).[陈耀丹,王连明.基于卷积神经网络的人脸识别方法[J].东北师大学报(自然科学版),2016,48(2):70-76.]
[3]Lecun Y,Bottou L,Bengio Y,et al.Gradient-based learning applied to document recognition[J].Proceedings of the IEEE,1998,86(11):2278-2324.
[4]CHEN Rui,LIN Da.Key point location of face based on cascaded convolution neural network[J].Journal of Sichuan University of Science&Engineering(Natural Science Editton),2017,30(1):32-37(in Chinese).[陈锐,林达.基于级联卷积神经网络的人脸关键点定位[J].四川理工学院学报(自然科学版),2017,30(1):32-37.]
[5]HE Pengcheng.Improved convolution neural network model and its application[D].Dalian:Dalian University of Technology,2015(in Chinese).[何鹏程.改进的卷积神经网络模型及其应用研究[D].大连:大连理工大学,2015.]
[6]XU Xiao,CHEN Yang,ZHANG Feiyun,et al.Research on iris localization based on deep deconvolution network[J].Journal of Integration Technology,2016,5(1):57-67(in Chinese).[徐霄,陈阳,张飞云,等.基于深度反卷积网络的虹膜定位方法研究[J].集成技术,2016,5(1):57-67.]
[7]HE Xin,CHEN Xun.FVR based on deep learning and improved Gabor feature fusion[J].Computer Simulation,2018,35(11):356-361(in Chinese).[何鑫,陈迅.基于深度学习与改进Gabor特征融合的FVR[J].计算机仿真,2018,35(11):356-361.]
[8]LIU Wanjun,LIANG Xuejian,QU Haicheng.Study on learning performance of convolution neural networks with different pooling models[J].Journal of Image and Graphics,2016,21(9):1178-1190(in Chinese).[刘万军,梁雪剑,曲海成.不同池化模型的卷积神经网络学习性能研究[J].中国图象图形学报,2016,21(9):1178-1190.]
[9]WANG Zeying.Research on block game verification code and verification code recognition based on small sample[D].Xiangtan:Xiangtan University,2016(in Chinese).[王泽颖.滑块游戏验证码和基于小样本的验证码识别研究[D].湘潭:湘潭大学,2016.]
[10]XUE Haotian,YANG Jingdong,TAN Kaide.An improved BP neural network for handwriting recognition[J].Electronic Science and Technology,2015,28(5):20-23(in Chinese).[薛皓天,杨晶东,谈凯德.一种改进的BP神经网络在手写体识别上的应用[J].电子科技,2015,28(5):20-23.]
[2]CHEN Yaodan, WANG Lianming.Face recognition method based on convolution neural network[J].Journal of Northeast Normal University(Natural Science),2016,48(2):70-76(in Chinese).[陈耀丹,王连明.基于卷积神经网络的人脸识别方法[J].东北师大学报(自然科学版),2016,48(2):70-76.]
[3]Lecun Y,Bottou L,Bengio Y,et al.Gradient-based learning applied to document recognition[J].Proceedings of the IEEE,1998,86(11):2278-2324.
[4]CHEN Rui,LIN Da.Key point location of face based on cascaded convolution neural network[J].Journal of Sichuan University of Science&Engineering(Natural Science Editton),2017,30(1):32-37(in Chinese).[陈锐,林达.基于级联卷积神经网络的人脸关键点定位[J].四川理工学院学报(自然科学版),2017,30(1):32-37.]
[5]HE Pengcheng.Improved convolution neural network model and its application[D].Dalian:Dalian University of Technology,2015(in Chinese).[何鹏程.改进的卷积神经网络模型及其应用研究[D].大连:大连理工大学,2015.]
[6]XU Xiao,CHEN Yang,ZHANG Feiyun,et al.Research on iris localization based on deep deconvolution network[J].Journal of Integration Technology,2016,5(1):57-67(in Chinese).[徐霄,陈阳,张飞云,等.基于深度反卷积网络的虹膜定位方法研究[J].集成技术,2016,5(1):57-67.]
[7]HE Xin,CHEN Xun.FVR based on deep learning and improved Gabor feature fusion[J].Computer Simulation,2018,35(11):356-361(in Chinese).[何鑫,陈迅.基于深度学习与改进Gabor特征融合的FVR[J].计算机仿真,2018,35(11):356-361.]
[8]LIU Wanjun,LIANG Xuejian,QU Haicheng.Study on learning performance of convolution neural networks with different pooling models[J].Journal of Image and Graphics,2016,21(9):1178-1190(in Chinese).[刘万军,梁雪剑,曲海成.不同池化模型的卷积神经网络学习性能研究[J].中国图象图形学报,2016,21(9):1178-1190.]
[9]WANG Zeying.Research on block game verification code and verification code recognition based on small sample[D].Xiangtan:Xiangtan University,2016(in Chinese).[王泽颖.滑块游戏验证码和基于小样本的验证码识别研究[D].湘潭:湘潭大学,2016.]
[10]XUE Haotian,YANG Jingdong,TAN Kaide.An improved BP neural network for handwriting recognition[J].Electronic Science and Technology,2015,28(5):20-23(in Chinese).[薛皓天,杨晶东,谈凯德.一种改进的BP神经网络在手写体识别上的应用[J].电子科技,2015,28(5):20-23.]