[ Home ] [ Up ] [ S-Cell ]

The Simplicial-CNN

If we use the Simplicial Canonical PWL representation that I developed in my thesis to design the cell of a CNN, then we obtain the so-called Simplicial CNN or S-CNN (also baptized in Andreou lab as the Super- CNN) for short. This is a universal CNN which in addition can be implemented using a very efficient and fully programmable mixed-signal circuit.  In addition, it has several numerical interesting features. We have included some results in Image processing of Gray-Level and color images. 

       3D animated new signs Visit the Gallery of Images

       3D animated new signs Get a tutorial paper on  the S-CNN

Multinested CNN

The multinested CNN uses a multiple- nested discriminant PWL function to produce Boolean functions with a minimal complexity. Right now we are working in the synthesis of all templates for Von Neumann neighborhoods. 

Publications

  •  P. Julián, R. Dogaru, L. O. Chua, ``A Piecewise-Linear Simplicial Coupling Cell for CNN Gray-Level Image Processing,'' IEEE Trans. Circuits and Systems, Vol. 49, No. 7, pp. 904-913, July 2002.

  •  R. Dogaru, P. Julián, Leon O. Chua, Manfred Glesner, "The Simplicial Neural Cell and its Mixed-Signal Circuit Implementation: An Efficient Neural Network Architecture for Intelligent Signal Processing in Portable Multimedia Applications," IEEE Transactions on Neural Networks, Vol. 13, No. 4, pp. 995-1008, July 2002.

  •  P. Julián, R. Dogaru, L. O. Chua, ``Discrete Time Cellular Neural Networks: A Simplicial Piecewise Linear Approach,'' Proc. of the IEEE Workshop on Nonlinear Signal and Image Processing, NSIP 2001, June 3-6, Baltimore, Maryland, US.

  •  P. Julián, R. Dogaru, M. Haenggi, L. O. Chua, ``An Algorithm for the Search of Multinested Cellular Neural Networks Parameters,'' accepted in the IEEE South-American Workshop on Circuits and Systems, SAWCAS 2001, Nov. 26-30, Rio de Janeiro, Brasil and Buenos Aires, Argentina.

  •  P. Julián, R. Dogaru, L. O. Chua, ``A Piecewise-Linear Simplicial Coupling Cell for CNN Gray-Level Image Processing,'' in Proc. of IEEE ISCAS 2001. 

  •  R. Dogaru, P. Julián, L. O. Chua, ``A Robust and Efficient CNN Cell Circuit Using Simplicial Neuro-fuzzy Inferences for Fast Image Processing,'' in Proc. IEEE ISCAS 2001.