Abstract
The hierarchical convolutional neural network models are considered promising for robust object detection/recognition. These models require huge computational power for performing a large number of multiply-and-accumulation (MAC) operations. In this paper, first we discuss efficient calculation schemes suitable for 2D MAC operations. Then we review the related algorithms and LSI architecture proposed in our previous work, in which we use a projection-field-type network architecture with sorting of neuron outputs by magnitude. For the LSI implementation, we adopt a merged/mixed analog-digital circuit approach using a large number of analog or pulse modulation circuits. We demonstrate the validity of our LSI architecture by testing proof-of-concept LSIs. It is essential to develop efficient and parallel A/D and D/A conversion circuits in order to connect a lot of on-chip analog circuits with the external digital system. In this paper, we also propose such an A/D conversion circuit scheme.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Fukushima, K., Miyake, S.: Neocognitron: A new algorithm for pattern recognition tolerant of deformations and shifts in position. Pattern Recognition 15, 455–469 (1982)
Lawrence, S., Giles, C.L., Tsoi, A.C., Back, A.D.: Face recognition: A convolutional neural-network approach. IEEE Trans. Neural Networks 8(1), 98–113 (1997)
Neubauer, C.: Evaluation of convolutional neural networks for visual recognition. IEEE Trans. Neural Networks 9, 685–696 (1998)
Matsugu, M., Mori, K., Ishii, M., Mitarai, Y.: Convolutional spiking neural network model for robust face detection. In: Proc. Int. Conf. on Neural Information Processing (ICONIP), pp. 660–664 (2002)
Kyo, S., Koga, T., Okazaki, S., Uchida, R., Yoshimoto, S., Kuroda, I.: A 51.2GOPS scalable video recognition processor for intelligent cruise control based on a linear array of 128 4-way VLIW processing elements. In: IEEE Int. Solid-State Circuits Conf. Dig., pp. 48–49 (2003)
Nakajima, M., Noda, H., Dosaka, K., Nakata, K., Higashida, M., Yamamoto, O., Mizumoto, K., Kondo, H., Shimazu, Y., Arimoto, K., Saitoh, K., Shimizu, T.: A 40GOPS 250mW massively parallel processor based on matrix architecture. In: IEEE Int. Solid-State Circuits Conf. Dig., pp. 410–411 (2006)
Boser, B.E., Säckinger, E., Bromley, J., Cun, Y.L., Jackel, L.D.: An analog neural network processor with programmable topology. IEEE J. Solid-State Circuits 26(12), 2017–2025 (1991)
Korekado, K., Morie, T., Nomura, O., Ando, H., Nakano, T., Matsugu, M., Iwata, A.: A VLSI convolutional neural network for image recognition using merged/mixed analog-digital architecture. J. Intelligent & Fuzzy Systems 15(3/4), 173–179 (2004)
Korekado, K., Morie, T., Nomura, O., Nakano, T., Matsugu, M., Iwata, A.: An image filtering processor for face/object recognition using merged/mixed analog-digital architecture. In: Symposium on VLSI Circuits, Digest of Technical papers, Kyoto 222–223 (2005)
Nomura, O., Morie, T., Korekado, K., Nakano, T., Matsugu, M., Iwata, A.: An image-filtering LSI processor architecture for face/object recognition using a sorted projection-field model based on a merged/mixed analog-digital architecture. IEICE Trans. Electron. E89-C(6), 781–791 (2006)
Sakai, M., Morie, T., Mitarai, M., Korekado, K.: Design of an 2D image matching processor LSI based on merged analog/digital architecture. In: RISP International Workshop on Nonlinear Circuits and Signal Processing (NCSP 2007), Shanghai, China, pp. 81–84 (2007)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2008 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Nomura, O., Morie, T. (2008). Projection-Field-Type VLSI Convolutional Neural Networks Using Merged/Mixed Analog-Digital Approach . In: Ishikawa, M., Doya, K., Miyamoto, H., Yamakawa, T. (eds) Neural Information Processing. ICONIP 2007. Lecture Notes in Computer Science, vol 4984. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-69158-7_111
Download citation
DOI: https://doi.org/10.1007/978-3-540-69158-7_111
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-69154-9
Online ISBN: 978-3-540-69158-7
eBook Packages: Computer ScienceComputer Science (R0)