Skip to main content
SpringerLink
Log in

A Mixed Mode Neural Network Circuitry for Object Recognition Application

  • Published:
Circuits, Systems, and Signal Processing Aims and scope Submit manuscript

Abstract

A general purpose Conic Section Function Neural Network (CSFNN) circuitry in Very Large Scale Integration (VLSI) has been designed for an object recognition application. CSFNN is capable of making open and closed decision regions by combining the propagation rules of Radial Basis Functions (RBF) and Multilayer Perceptrons (MLP) on a single neural network with a unique propagation rule. Chip-in-the-loop learning technique was used during the training process. Utilizing mixed-mode hardware techniques, the inputs of the network and the feedforward signals are all analog while the control unit and storage of the network parameters are fully digital. CSFNN circuitry architecture is problem independent and consists of 16 inputs, 16 hidden layer neurons and 8 outputs. Inheriting the merits of CSFNN, the circuitry has good recognition performance on several objects with invariance to pose, lighting, and brightness. The designed hardware achieved a good recognition performance by means of both accuracy and computational time comparable to CSFNN software.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15
Fig. 16
Fig. 17

Similar content being viewed by others

References

  1. J. Baker, CMOS Circuit Design, Layout, and Simulation, 2nd edn. (Wiley/IEEE Press, New York, 1997)

    Google Scholar 

  2. S. Bibyk, M. Ismail, T. Borgstrom, K. Adkins, R. Kaul, N. Khachab, S. Dupuie, Current-mode neural network building blocks for analog MOS VLSI, in IEEE Int. Symp. on Circuits and Systems, vol. 4 (1990), pp. 3283–3285

    Chapter  Google Scholar 

  3. G.M. Bo, D.D. Caviglia, M. Valle, A current mode CMOS multi-layer perceptron chip, in Proc. of 5th Int. Conf. on Microelectronics for Neural Networks (1996), pp. 103–106

    Chapter  Google Scholar 

  4. G. Cairns, L. Tarassenko, Learning with analogue VLSI MLPs, in Proc. of the 4th Int. Conf. on Microelectronics for Neural Networks and Fuzzy Systems (1994), pp. 67–76

    Chapter  Google Scholar 

  5. G. Cauwenberghs, M.A. Bayoumi, Learning on Silicon Adaptive VLSI Neural Systems (Kluwer Academic, Norwell, 1999)

    MATH  Google Scholar 

  6. J. Chen, T. Shibata, A neuron-MOS-based VLSI implementation of pulse-coupled neural networks for image feature generation. IEEE Trans. Circuits Syst. I, Regul. Pap. 57, 1143–1153 (2010)

    Article  MathSciNet  Google Scholar 

  7. W. Chung-Yu, C. Chiu-Hung, A learnable cellular neural network structure with ratio memory for image processing. IEEE Trans. Circuits Syst. I, Fundam. Theory Appl. 49, 1713–1723 (2002)

    Article  Google Scholar 

  8. S. Cohen, N. Intrator, Automatic model selection in a hybrid perceptron/radial network, in Multiple Classifier Systems. Lecture Notes in Computer Science, vol. 2096 (2001), pp. 440–454

    Chapter  Google Scholar 

  9. G. Dorffner, Unified frameworks for MLP and RBFNs: introducing conic section function networks. Cybern. Syst. 25, 511–554 (1994)

    Article  MathSciNet  Google Scholar 

  10. B. Erkmen, T. Yildirim, Obtaining decision boundaries of CSFNN neurons using current mode analog circuitry, in Proc. of the 18th European Conf. on Circuit Theory and Design (2007), pp. 807–810

    Google Scholar 

  11. B. Erkmen, Mixed mode hardware realization of a general purposed artificial neural network. Ph.D. Thesis, Yildiz Technical University, Turkey (2007)

  12. B. Erkmen, N. Kahraman, R.A. Vural, T. Yildirim, Conic section function neural network circuitry for offline signature recognition. IEEE Trans. Neural Netw. 21(4), 667–672 (2010)

    Article  Google Scholar 

  13. H. Esmaelzadeh, C. Farshbaf, S.M. Lucas, S.M. Fakhraie, Digital implementation for conic section function networks, in Proc. of the 16th Int. Conf. on Microelectronics (2004), pp. 564–567

    Google Scholar 

  14. S. Haykin, Neural Networks. A Comprehensive Foundation (Macmillan, New York, 1994), pp. 363–370

    MATH  Google Scholar 

  15. D.F. Hoeschele, Analog-to-Digital and Digital-to-Analog Conversion Techniques, 2nd edn. (Wiley, New York, 1994)

    Google Scholar 

  16. J.-Y. Kim et al., A 201.4 GOPS 496 mW real-time multi-object recognition processor with bio-inspired neural perception engine. IEEE J. Solid-State Circuits 45(1), 32–45 (2010)

    Article  Google Scholar 

  17. K.J. Kyoung, Fully-pipelining hardware implementation of neural network for text-based images retrieval, in Lecture Notes in Computer Science, vol. 3973 (2006), pp. 1350–1356

    Google Scholar 

  18. Y. Lecun, F.J. Huang, L. Bottou, Learning methods for generic object recognition with invariance to pose and lighting, in Proc. of the Computer Vision and Pattern Recognition Conference, vol. 2 (2004), pp. 97–104

    Google Scholar 

  19. G. Lowe, Distinctive image features from scale-invariant keypoints. Int. J. Comput. Vis. 60, 91–110 (2004)

    Article  Google Scholar 

  20. C. Lu, B. Shi, L. Chen, A programmable on-chip BP learning neural network with enhanced neuron characteristics, in IEEE Int. Symp. on Circuits and Systems, May 2001, vol. 3 (2001), pp. 573–576

    Google Scholar 

  21. P. Masa, K. Hoen, H. Wallinga, 70 input, 20 nanosecond pattern classifier, in IEEE Int. Conf. on Neural Networks (1994), pp. 1854–1859

    Google Scholar 

  22. M. Mirhassani, M. Ahmadi, W.C. Miller, A mixed-signal VLSI neural network with on-chip learning, in Canadian Conf. on Electrical and Computer Engineering (2003), pp. 1–4

    Google Scholar 

  23. M. Mirhassani, M. Ahmadi, W.C. Miller, A feed-forward time-multiplexed neural network with mixed-signal neuron-synapse arrays, in The Microelectronic Engineering Journal, vol. 84 (2007), pp. 300–307. Special issue on VLSI Design and Test

    Google Scholar 

  24. A.J. Montalvo, R.S. Gyurcsik, J.J. Paulos, Toward a general-purpose analog VLSI neural network with on-chip learning. IEEE Trans. Neural Netw. 8, 413–423 (1997)

    Article  Google Scholar 

  25. P. Pavan, R. Bez, P. Olivo, E. Zanoni, Flash memory cells: an overview, in Proceedings of the IEEE, vol. 85 (1997), pp. 1248–1271

    Google Scholar 

  26. A. Schmid, Y. Leblebici, D. Mlynek, Mixed analogue-digital artificial-neural-network architecture with on-chip learning, in IEE Proc. on Circuits, Devices and Systems, vol. 146 (1999), pp. 345–349

    Google Scholar 

  27. A.K. Sharma, Advanced Semiconductor Memories, Architectures and Applications (Wiley-Interscience/IEEE Press, Norwell, 2003)

    Google Scholar 

  28. B.J. Sheu, J. Choi, Neural Information Processing and VLSI (Kluwer Academic, New York, 1995), pp. 3–16

    Book  MATH  Google Scholar 

  29. F. Stupmann, S. Rode, G. Geske, Single chip-VLSI-realization of a neural net for image recognition, in IEEE Int. Conf. on Industrial Technology, vol. 2 (2002), pp. 1345–1348

    Google Scholar 

  30. R.A. Vural, N. Kahraman, B. Erkmen, T. Yildirim, Object recognition on general purposed conic section function neural network integrated circuit, in 16th IEEE Signal Processing, Communication and Applications Conference (2008), pp. 1–4

    Google Scholar 

  31. S. Wolpert, L.A. Lee, J.F. Heisler, Circuits for a VLSI-based standalone backpropagation neural network, in Proc. of the IEEE Bioengineering Conference Northeast (1992), pp. 47–48

    Google Scholar 

  32. M.P. Yang, Implementation of an RBF neural network on embedded systems: real-time face tracking and identity verification. IEEE Trans. Neural Netw. 14, 1162–1175 (2003)

    Article  Google Scholar 

  33. T. Yıldırım, J.S. Marsland, An RBF/MLP hybrid neural network implemented in VLSI hardware, in Proc. 8th Int. Conf. on Neural Networks and Their Applications (1995), pp. 156–160

    Google Scholar 

Download references

Acknowledgement

This work was supported by the TUBITAK—The Scientific and Technological Research Council of Turkey. Project Number: 104E133.

Author information

Authors and Affiliations

  1. Department of Electronics and Communication Engineering, Yildiz Technical University, 34220, Istanbul, Turkey

    Burcu Erkmen, Revna Acar Vural, Nihan Kahraman & Tulay Yildirim

Authors
  1. Burcu Erkmen
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Revna Acar Vural
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Nihan Kahraman
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Tulay Yildirim
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Burcu Erkmen.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Erkmen, B., Vural, R.A., Kahraman, N. et al. A Mixed Mode Neural Network Circuitry for Object Recognition Application. Circuits Syst Signal Process 32, 29–46 (2013). https://doi.org/10.1007/s00034-012-9458-2

Download citation

  • Received:

  • Revised:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00034-012-9458-2

Keywords

Search

Navigation