The FPGA prototyping of Iris recognition for biometric identification employing neural network

F. Mohd-Yasin, A. L. Tan, Md. Mamun Ibne Reaz

Research output: Chapter in Book/Report/Conference proceedingConference contribution

47 Citations (Scopus)

Abstract

In this paper, we present the realization of of IRIS recognition for biometric identification employing neural network on Altera FLEX10K FPGA device that allows for efficient hardware implementation. This method consists of two main parts, which are image processing and recognition. Image processing is implemented by using MATLAB and back propagation method was used for recognition. The iris recognition neural network architecture comprises of three layers: input layer with three neurons, hidden layer with two neurons and output layer with one neuron. Sigmoid transfer function is used for both hidden layer and output layer neurons. The timing analysis for the validation, functionality and performance of the model is performed using Aldec Active HDL and the logic synthesis was performed using Synplify. Iris vector from captured human iris has been used to validate the effectiveness of the model. Test on the sample of 100 data showed an accuracy of 88.6% in recognizing the sample of irises.

Original languageEnglish
Title of host publicationProceedings of the International Conference on Microelectronics, ICM
Pages458-461
Number of pages4
Publication statusPublished - 2004
Externally publishedYes
Event16th International Conference on Microelectronics, ICM 2004 - Tunis
Duration: 6 Dec 20048 Dec 2004

Other

Other16th International Conference on Microelectronics, ICM 2004
CityTunis
Period6/12/048/12/04

Fingerprint

Biometrics
Neurons
Field programmable gate arrays (FPGA)
Neural networks
Image processing
Image recognition
Network architecture
Backpropagation
MATLAB
Transfer functions
Hardware

Keywords

  • Artificial Intelligence
  • Biometric Identification
  • FPGA Prototyping
  • Iris
  • Neural Network

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Mohd-Yasin, F., Tan, A. L., & Ibne Reaz, M. M. (2004). The FPGA prototyping of Iris recognition for biometric identification employing neural network. In Proceedings of the International Conference on Microelectronics, ICM (pp. 458-461)

The FPGA prototyping of Iris recognition for biometric identification employing neural network. / Mohd-Yasin, F.; Tan, A. L.; Ibne Reaz, Md. Mamun.

Proceedings of the International Conference on Microelectronics, ICM. 2004. p. 458-461.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Mohd-Yasin, F, Tan, AL & Ibne Reaz, MM 2004, The FPGA prototyping of Iris recognition for biometric identification employing neural network. in Proceedings of the International Conference on Microelectronics, ICM. pp. 458-461, 16th International Conference on Microelectronics, ICM 2004, Tunis, 6/12/04.
Mohd-Yasin F, Tan AL, Ibne Reaz MM. The FPGA prototyping of Iris recognition for biometric identification employing neural network. In Proceedings of the International Conference on Microelectronics, ICM. 2004. p. 458-461
Mohd-Yasin, F. ; Tan, A. L. ; Ibne Reaz, Md. Mamun. / The FPGA prototyping of Iris recognition for biometric identification employing neural network. Proceedings of the International Conference on Microelectronics, ICM. 2004. pp. 458-461
@inproceedings{01dc1cf547db46158a2f8ca2330559fd,
title = "The FPGA prototyping of Iris recognition for biometric identification employing neural network",
abstract = "In this paper, we present the realization of of IRIS recognition for biometric identification employing neural network on Altera FLEX10K FPGA device that allows for efficient hardware implementation. This method consists of two main parts, which are image processing and recognition. Image processing is implemented by using MATLAB and back propagation method was used for recognition. The iris recognition neural network architecture comprises of three layers: input layer with three neurons, hidden layer with two neurons and output layer with one neuron. Sigmoid transfer function is used for both hidden layer and output layer neurons. The timing analysis for the validation, functionality and performance of the model is performed using Aldec Active HDL and the logic synthesis was performed using Synplify. Iris vector from captured human iris has been used to validate the effectiveness of the model. Test on the sample of 100 data showed an accuracy of 88.6{\%} in recognizing the sample of irises.",
keywords = "Artificial Intelligence, Biometric Identification, FPGA Prototyping, Iris, Neural Network",
author = "F. Mohd-Yasin and Tan, {A. L.} and {Ibne Reaz}, {Md. Mamun}",
year = "2004",
language = "English",
pages = "458--461",
booktitle = "Proceedings of the International Conference on Microelectronics, ICM",

}

TY - GEN

T1 - The FPGA prototyping of Iris recognition for biometric identification employing neural network

AU - Mohd-Yasin, F.

AU - Tan, A. L.

AU - Ibne Reaz, Md. Mamun

PY - 2004

Y1 - 2004

N2 - In this paper, we present the realization of of IRIS recognition for biometric identification employing neural network on Altera FLEX10K FPGA device that allows for efficient hardware implementation. This method consists of two main parts, which are image processing and recognition. Image processing is implemented by using MATLAB and back propagation method was used for recognition. The iris recognition neural network architecture comprises of three layers: input layer with three neurons, hidden layer with two neurons and output layer with one neuron. Sigmoid transfer function is used for both hidden layer and output layer neurons. The timing analysis for the validation, functionality and performance of the model is performed using Aldec Active HDL and the logic synthesis was performed using Synplify. Iris vector from captured human iris has been used to validate the effectiveness of the model. Test on the sample of 100 data showed an accuracy of 88.6% in recognizing the sample of irises.

AB - In this paper, we present the realization of of IRIS recognition for biometric identification employing neural network on Altera FLEX10K FPGA device that allows for efficient hardware implementation. This method consists of two main parts, which are image processing and recognition. Image processing is implemented by using MATLAB and back propagation method was used for recognition. The iris recognition neural network architecture comprises of three layers: input layer with three neurons, hidden layer with two neurons and output layer with one neuron. Sigmoid transfer function is used for both hidden layer and output layer neurons. The timing analysis for the validation, functionality and performance of the model is performed using Aldec Active HDL and the logic synthesis was performed using Synplify. Iris vector from captured human iris has been used to validate the effectiveness of the model. Test on the sample of 100 data showed an accuracy of 88.6% in recognizing the sample of irises.

KW - Artificial Intelligence

KW - Biometric Identification

KW - FPGA Prototyping

KW - Iris

KW - Neural Network

UR - http://www.scopus.com/inward/record.url?scp=21644487177&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=21644487177&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:21644487177

SP - 458

EP - 461

BT - Proceedings of the International Conference on Microelectronics, ICM

ER -