基于ARM和深度学习的大数据指纹识别系统设计
|
摘要
针对传统指纹识别系统在面对大数据指纹图像时具有识别效率不高、需要手动设计提取的特征的缺点,提出了一种基于ARM和深度学习的大数据指纹识别系统.首先,描述了指纹识别系统的原理图.然后,设计了系统硬件框图,采用S3C2410C作为微处理器,采用FPS200指纹图像作为传感器,并设计了两者之间的接口电路;最后,重点设计了指纹识别的软件过程,建立一个可以进行指纹自动识别的通用多层深层神经网络.通过设计系统软硬件并进行测试,结果表明文中设计的指纹识别系统具有很高的指纹识别准确度,能有效处理大数据指纹图像的识别,且与其他基于人工提取特征的方法相比,具有更高的识别正确率和识别效率.
Aiming at the traditional fingerprint recognition system had the problems such as not high recognition rate and having to design the feature manually,a big data fingerprint recognition system based on ARM and deep learning was proposed. Firstly,the fingerprint system was specified. Then,the S3 C2410 C was used as the micro-processor and FPS200 as the fingerprint sensor,the interface circuit between them was also designed. Finally,the software process of fingerprint recognition was designed in detail,the input layer,convolutional layer,feature automatically extraction layer and the output layer was constructed to be a deep neural network. Via designing the system hardware and software and the test,the result shows the method has the high fingerprint recognition rate,which achieve the fingerprint figure for big data. Compared with the other methods which uses the feature extracted manually,the method has higher recognition rate and recognition efficiency.
Aiming at the traditional fingerprint recognition system had the problems such as not high recognition rate and having to design the feature manually,a big data fingerprint recognition system based on ARM and deep learning was proposed. Firstly,the fingerprint system was specified. Then,the S3 C2410 C was used as the micro-processor and FPS200 as the fingerprint sensor,the interface circuit between them was also designed. Finally,the software process of fingerprint recognition was designed in detail,the input layer,convolutional layer,feature automatically extraction layer and the output layer was constructed to be a deep neural network. Via designing the system hardware and software and the test,the result shows the method has the high fingerprint recognition rate,which achieve the fingerprint figure for big data. Compared with the other methods which uses the feature extracted manually,the method has higher recognition rate and recognition efficiency.
引文
[1]Maltoni D,Maio D,Jain A K,et al. Handbook of fingerprint recognition[M]. New York:Springer Verlag,2003:1-9.
[2]钦小平,潘炼.基于ARM的QNMV指纹识别系统的研究[J].计算机测量与控制,2012,20(4):1076-1078.
[3]Fons M,Fons F,Canto E. Design of an embedded fingerprint matcher system[C]//St. Petersburg:IEEE Tenth International Symposium on Consumer Electronics,2006:1-6.
[4]傅磊,史延东,李飞,等.一种新型指纹识别系统的设计与实现[J].电子技术应用,2010(2):16-18.
[5]Fu Q,Mansjur D S,Juang B H. Empirical system learning for statistics pattern recognition with non-uniform error criteria[J].Signal Processing,2010,58(9):4621-4633.
[6]孙萍.基于AT89S52的指纹识别系统的设计[J].电气应用,2010,29(9):74-80.
[7]陈书谦,傅扬烈.改进的指纹识别算法嵌入式应用研究[J].微电子学与计算机,2012(10):181-184.
[8]韩智,刘昌平.基于多种特征融合的指纹识别方法[J].计算机科学,2010,37(7):254-259.
[9]许沪敏,杨森,朱涛.基于Delaunay网格和DLP的三维指纹识别系统设计[J].计算机测量与控制,2014,22(10):3429-3432.
[10]Deng L,Yu D. Deep learning for signal and information processing[J]. Microsoft Research monograph,2013.
[11]Bengio Y,Courville A. Deep learning of representations[M]. Handbook on Neural Information Processing. Springer Berlin Heidelberg,2013.
[12]杨钊,陶大鹏,张树业,等.大数据下的基于深度神经网的相似汉字识别[J].通信学报,2014,35(9):184-189.
[13]张玉清,董颖,柳彩云,等.深度学习应用于网络空间安全的现状、趋势与展望[J].计算机研究与发展,2018,55(6):1117-1142.
[14]林付春,刘宇红,张达峰,等.基于深度学习的智能路牌识别系统设计[J].电子技术应用,2018,480(6):74-77.
[2]钦小平,潘炼.基于ARM的QNMV指纹识别系统的研究[J].计算机测量与控制,2012,20(4):1076-1078.
[3]Fons M,Fons F,Canto E. Design of an embedded fingerprint matcher system[C]//St. Petersburg:IEEE Tenth International Symposium on Consumer Electronics,2006:1-6.
[4]傅磊,史延东,李飞,等.一种新型指纹识别系统的设计与实现[J].电子技术应用,2010(2):16-18.
[5]Fu Q,Mansjur D S,Juang B H. Empirical system learning for statistics pattern recognition with non-uniform error criteria[J].Signal Processing,2010,58(9):4621-4633.
[6]孙萍.基于AT89S52的指纹识别系统的设计[J].电气应用,2010,29(9):74-80.
[7]陈书谦,傅扬烈.改进的指纹识别算法嵌入式应用研究[J].微电子学与计算机,2012(10):181-184.
[8]韩智,刘昌平.基于多种特征融合的指纹识别方法[J].计算机科学,2010,37(7):254-259.
[9]许沪敏,杨森,朱涛.基于Delaunay网格和DLP的三维指纹识别系统设计[J].计算机测量与控制,2014,22(10):3429-3432.
[10]Deng L,Yu D. Deep learning for signal and information processing[J]. Microsoft Research monograph,2013.
[11]Bengio Y,Courville A. Deep learning of representations[M]. Handbook on Neural Information Processing. Springer Berlin Heidelberg,2013.
[12]杨钊,陶大鹏,张树业,等.大数据下的基于深度神经网的相似汉字识别[J].通信学报,2014,35(9):184-189.
[13]张玉清,董颖,柳彩云,等.深度学习应用于网络空间安全的现状、趋势与展望[J].计算机研究与发展,2018,55(6):1117-1142.
[14]林付春,刘宇红,张达峰,等.基于深度学习的智能路牌识别系统设计[J].电子技术应用,2018,480(6):74-77.