CNNA 2004 Program

 

Thursday, July 22,  09:00-10:30

Session P1: Plenary talk on Nanotechnology


09:00-09:40            Emerging Nanoelectronic Device and Circuit Technologies

                               Wolfgang Porod and Ádám Csurgay

                               University of Notre Dame, United States; Pazmany P. University, Budapest

 

09:40-10:20            Image Processing by RTD-based CNN

                               Pinaki Mazumder

                               University of Michigan, United States

 

Thursday, July 22,  10:40-12:20

Session I1: Implementation 1: CNN computers, software simulators, FPGA emulation

10:40-11:00  

Bi-i: a Standalone Cellular Vision System, Part II. Topographic and Non-topographic Algorithms and Related Applications

Csaba Rekeczky and Ákos Zarándy

MTA-SZTAKI, Hungary

11:00-11:20 

Bi-i: a Standalone Cellular Vision System, Part I. Architecture and Ultra High Frame Rate Processing Examples

Ákos Zarándy and Csaba Rekeczky

MTA SZTAKI, Hungary

11:20-11:40  

E^3: Extended CNN Paradigm for Emergent Dynamical System Emulation

Riccardo Caponetto, Stefano Fazzino, Luigi Fortuna and Mattia Frasca

University of Catania, Italy

11:40-12:00  

CELL: a New Interactive Tool for CNN Algorithms Implementation

Mario Salerno, Daniele Casali, Giovanni Costantini and Massimo Carota

University of Rome "Tor Vergata" - Department of, Italy

12:00-12:20  

Low-cost CNN-UM GAPU implementation on FPGA

László Beke, Zoltán Nagy and Péter Szolgay

University of Veszprém, Hungary

 

Session T1: Theory 1: Linear filters and chaotic circuits

10:20-10:40  

On the Sensitivity of CNN Linear Spatial Filters: Homogeneous Template Variations

Emilian David, Romeo Ghinea, Michael Ansorge and Liviu Goras

"Gh. Asachi" Technical University of Iasi, Romania; University of Neuchatel, Switzerland

10:40-11:00 

On the Sensitivity of CNN Linear Spatial Filters: Non-Homogeneous Template Variations

Emilian David, Michael Ansorge, Liviu Goras and Victor Grigoras

"Gh. Asachi" Technical University of Iasi, Romania; University of Neuchatel, Switzerland

11:00-11:20  

Generation of New Hyperchaotic Multi-scroll Attractors in three-cell CNN's

Donato Cafagna and Giuseppe Grassi

Universita degli Studi di Lecce, Italy

11:20-12:00  

Dynamics of Cellular Neural Networks: Occurrence of the Chaotic Beats Phenomenon

Donato Cafagna and Giuseppe Grassi

Universita degli Studi di Lecce, Italy

12:00-12:20  

Complex Behavior in a Ring of Chaotic Circuits Related with Intermittency

Yoko Uwate and Yoshifumi Nishio

Tokushima University, Japan

 

 

Thursday, July 22,  13:40-15:00

Session I2: Implementation 2:

13:40-14:00  

Evolution of the Programmable Optical Array Computer (POAC)

Ahmed Ayoub, Szabolcs Tőkés, László Orzó and Tamás Roska

MTA SZTAKI, Hungary

14:00-14:20  

Laptop POAC: a Compact Optical Implementation of CNN-UM

Szabolcs Tőkés, László Orzó, Ahmad Ayoub and Tamás Roska

Hungary; Egypt; Hungary

14:20-14:40  

Current Controlled Resistive Fuse Implementation in a Mixed-Mode Array Processor Core

Jonne Poikonen and Ari Paasio

University of Turku, Finland

14:40-15:00   

Pulse Mode CNN Image Processing System

Daisuke Maruyama and Hiroomi Hikawa

Oita University, Japan

 

Session T2: Theory 2:

13:40-14:00  

On Solitary and Periodic Waves in One-dimensional FitzHugh-Nagumo CNN's

Valeri Mladenov and Hans Hegt

the Netherlands

14:00-14:20  

On the Existence of Stable Equilibrium Points in Space Invariant Cellular Neural Networks

Marco Gilli, Mario Biey and Paolo Checco

Politecnico di Torino, Italy

14:20-15:00   

A New Computational Model for CNN-UMs and its Computational Complexity

Dávid Bálya, Gergely Timár, György Cserey and Tamás Roska

Analogic Lab., CARI-HAS, Hungary; Analogic Lab., CARI-HAS / PPKE, Hungary

15:00-15:20   

Toward Optimized Time-Multiplexed CNN Processing

László Kék

MTA SZTAKI, Hungary

 

 

Thursday, July 22,  15:20-16:40

Session I3: Implementation 3:

15:20-15:40

Analog Viterbi Decoder with Unilaternal Connections of the CNN

Son Hongrak, Kim Hyongsuk, Kim Hyunjung and Chua Leon

Chonbuk National University, Korea, Republic of; University of California, Berkeley, United States

15:40-16:00  

Fast and Efficient Implementation of Trigger-Wave Propagation on VLSI Cellular Processor Arrays

Piotr Dudek

University of Manchester (UMIST), United Kingdom

16:00-16:20  

Accuracy and Efficiency of Grey-level Image Filtering on VLSI Cellular Processor Arrays

Piotr Dudek

University of Manchester (UMIST), United Kingdom

16:20-16:40  

VLSI Implementation of a Binary CNN: First Measurement Results

Jacek Flak, Mika Laiho, Ari Paasio and Kari Halonen

Helsinki University of Technology, Finland; University of Turku, Finland

 

Session A1: Application 1:

15:20-15:40

Sorthest Path Problem with Pixel Level Snakes: Application to Robot Path Planning

David Vilarino and Csaba Rekeczky

University of Santiago de Compostela, Spain; Hungarian Academy of Sciences, Hungary

15:40-16:00  

Image Recognition on DTCNN with the Learned Structuring Elements of Hit-or-miss Operator

Kazuhiro Sumitomo and Akio Ushida

Takuma National College of Technology, Japan; Tokushima Bunri University, Japan

16:00-16:20  

Nonlinear Lifting Wavelet Scheme using Discrete Time Cellular Neural Networks for Lossless Image Coding

Hisashi Aomori, Kohei Kawakami, Tsuyoshi Otake, Nobuaki Takahashi and Mamoru Tanaka

Sophia University, Japan; IBM Japan, Ltd., Japan

16:20-16:40  

On the Design of a Class of Selective CNN Linear Filters

Radu Matei

Technical University of Iasi, Romania

 

Thursday, July 22,  17:00-18:00

Session D1: Panel discussion on Vision on CNN Technology

 


Friday, July 23,  09:00-09:50

Session P2: Plenary talk on Retina Modelling

        

            

 

09:00-09:50        Mammalian Retinal Vision: From Living Cells to Silicons

        

                           Botond Roska

 

                           Harvard University, United States

 

Friday, July 23,  10:20-12:00

Session T3: Theory 3

10:20-10:40  

Optimal Decomposition of Large-neighborhood CNN Templates

Mark ter Brugge, Jos Nijhuis and Lambert Spaanenburg

Dacolian b.v., Netherlands; University of Groningen, Netherlands; Lund University, Sweden

10:40-11:00 

Global Stability Analysis of Cellular Neural Networks with Multiple Time Delays

Sabri Arik

Istanbul University, Turkey

11:00-11:20  

Experimental Observation of Autowaves on the ACE16k CNN Chip

Mustak Erhan Yalcin, Johan A. K. Suykens and Joos Vandewalle

K.U.Leuven, ESAT/SCD, Belgium

11:20-11:40  

A Study with Respect to the Optimization of Cellular Neural Network Algorithms

Stephan Joseph Schwarz

University of Hannover, Germany

11:40-12:00 A Fast Analysis Method for both Coupled and Uncoupled Binary-Output Cloning Templates
Yohann Benedic, David Monnin and Jean Merckle
MIPS Lab. Haute Alsace University, France; French-German Research Institute of Saint Louis, France

 

Session A2: Application 2: Face and texture recognition

10:20-10:40  

Face analysis using CNN-UM

Zoltan Szlávik and Tamás Szirányi

Analogic and Neural Computing Laboratory, Hungary

10:40-11:00 

Fast Detection and Tracking of Faces in Uncontrolled Environments for Autonomous Robots Using the CNN-UM

John McRaven, Matthias Scheutz, György Cserey, Virgil Andronache and Wolfgang Porod

University of Notre Dame, United States

11:00-11:20  

Texture Decomposition with Cellular Neural Networks

Krzysztof Slot, Lukasz Kornatowski and Piotr Debiec

Institute of Electronics, TU of Lodz, Poland

11:20-11:40  

Analogic CNN Algorithms for Textile Quality Control Based on Optical and Tactile Sensory Inputs

Attila Kis, Ferenc Kovács and Péter Szolgay

Pazmany Peter Catholic Univ., Hungary; MTA SZTAKI, Hungary

11:40-12:00  

A Variational Method for Image Denoising, by using Cellular Neural Networks

Alexandru Gacsádi and Péter Szolgay

University of Oradea, Romania; Computer and Automation Institute, Hungary

 

 

Friday, July 23,  13:40-15:00

Session T4: Theory 4: Learning

13:40-14:00  

Associative Cellular Neural Networks with Adaptive Architecture

Oleksiy Dekhtyarenko, Alexander Reznik and Aleksander Sitchov

Institute of Mathematical Machines and Systems, Ukraine

14:00-14:20  

Data Mining for Nonlinear Prediction of Time Series by DT-CNN

Tatsuro Shimada, Yuta Watanabe, Kumiko Oshima, Nobuaki Takahashi, Tsuyoshi Otake and Mamoru Tanaka

Sophia University, Japan; IBM Japan, Ltd., Japan

14:20-14:40  

Cellular Neural Networks as Associative Memories Storing Grey-Scale Images

Giovanni Costantini, Daniele Casali and Massimo Carota

University of Rome "Tor Vergata", Italy

14:40-15:00   

Mapping propositional and multi-valued logic rule-based expert systems to the CNN-UM

Dávid Bálya

Analogic Lab., CARI-HAS, Hungary

 

Session A3: Application 3

13:40-14:00  

A Diagnostic Echocardiography System Boosted by CNN Technology 

Csaba Rekeczky, Binzberger Viktor, Hillier Daniel, Zsolt Czeilinger, Gergely Soós, György Cserey, László Kék and David Vilarino

MTA-SZTAKI, Hungary; Ppke, Hungary; GGy-Oki, Hungary; Decs-Usc, Spain

14:00-14:20  

Topographic Cellular Active Contour Techniques: Comparison of Different Approaches

Daniel Hillier, Viktor Binzberger, David Vilarino and Csaba Rekeczky

NIT - Computer and Automation Research Institute, Hungary; Dept. of Electronics, Univ. of Santiago de Compo, Spain; Jedlik Laboratories, ITK, Pázmány University, Hungary

14:20-14:40  

Symmetry feature extraction and "understanding"

László Havasi and Zoltán Szlávik

Department of Information Technology, PPCU, Hungary; Analogic and Neural Computing Systems Laboratory, Hungary

14:40-15:00   

Motion Estimation Matching Criterion Computation with Analog Circuit 

Lauri Koskinen, Mika Laiho, Ari Paasio and Kari Halonen

Helsinki University of Technology, Finland

 

 

 

 

 

Friday, July 23, 15:20-16:40

Session I4: Implementation 4

15:20-15:40  

CMOS Implementation of a Cellular Nonlinear Network for Image Segmentation

Simone Naso, Marco Storace, Giacomo Pruzzo and Mauro Parodi

DIBE- University of Genova, Italy

15:40-16:00

A Fuzzy Unit for a Mixed-Mode Array Processor

Laura Vesalainen, Jonne Poikonen and Ari Paasio

University of Turku, Finland

16:00-16:20  

On the One-Quadrant Template Design in a High Gain CNN Model

V.M. Brea, D.L. Vilarino, A. Paasio and D. Cabello

University of Santiago de Compostela, Spain; University of Turku, Finland

16:20-16:40  

Current Mode Analog Programmable CNN with Single-Polarity State Current

Mika Laiho, Ari Paasio and Kari Halonen

Helsinki University of Technology, Finland; University of Turku, Finland

 

Session A4: Application 4

15:20-15:40  

Real-Time Tracking Algorithm with Locking on a given Object for VLSI CNN-UM implementations

Samuel Xavier de Souza, Johan Suykens and Joos Vandewalle

K.U.Leuven, Belgium

15:40-16:00

A CNN image-compression algorithm for improved utilization of on-chip resources

László Orzó and Tamás Roska

MTA SZTAKI, Hungary

16:00-16:20  

High Dynamic Range Perception with Spatially Variant Exposure

Robert Wagner, Ákos Zarándy and Tamás Roska

Peter Pazmany Catholic University, Hungary; Analogic and Neural Computing Lab., MTA-SZTAKI, Hungary

16:20-16:40  

Bayesian Incorporation of Multiple Scales in Optical Flow Estimation

Levente Török, Ákos Zarándy and Tamás Roska

Analogic and Neural Computing Sys.,CARI-HAS, Hungary

 

 

 

Friday, July 23,  17:00-18:00

Session D2: Panel discussion on Physical Implementation and Application

 

 


Saturday, July 24,  09:00-10:40

Session S1: Special Session on Information Processing in Biological Systems

09:00-09:20  

Bioinspired CMOS Photosensor Adaption using Local Luminance Feedback 

Ricardo Carmona, J. Cuadri, C. Dominguez-Matas, Gustavo Linan and E. Roca

Instituto de Microelectronica de Sevilla, Spain

09:20-09:40 

CNNs and Motor Maps for Bio-inspired Collision Avoidance in Roving Robots 

Paolo Arena, Luigi Fortuna, Mattia Frasca, Roberto Pasqualino and Luca Patane

University of Catania, Italy

09:40-10:00  

Estimating Synchronization in Brain Electrical Activity from Epilepsy Patients with Cellular Neuronal 

R. Sowa, F. Mormann, A. Chernihovskyi, S. Florin, C.E.. Elger and K. Lehnertz

University Bonn, Germany

10:00-10:20  

Functional studies using the multichannel retinal simulator 

A. Lazr, Róbert Wagner, Dávid Bálya and Tamás Roska

MTA-SZTAKI, Hungary

10:20-10:40  

Novel Tools and Methods for Fast Identification of Emergent Behaviors in CNNs with Relevance to Biological Modeling 

Radu Dogaru

Polytechnic University of Bucharest, Romania

 

 

Session A5: Application 5

09:00-09:20  

CNN Algorithm for H.264 Motion Estimation Partitions 

Lauri Koskinen, Ari Paasio and Kari Halonen

Helsinki University of Technology, Finland

09:20-09:40 

Diffusion Analysis of Direction-Preserving Small-World CNN

Kazuya Tsuruta, Zonghuang Yang, Yoshifumi Nishio and Akio Ushida

Tokushima University, Japan; Tokushima Bunri University, Japan

09:40-10:00  

Polynomial CNN Models for PDE Based Image Multiscale Analysis

Fernando Corinto, Mario Biey and Marco Gilli

Politecnico di Torino, Italy

10:00-10:20  

Holistic Feature Extraction from Handwritten Words on Wave Computers

Kristóf Karacs and Tamás Roska

Hungarian Acadamy of Sciences, Hungary; Pazmany Peter Catholic University, Hungary

10:20-10:40  

Cellular Neural Networks To Model Innovation Systems

Pierpaolo Andriani, Francesca Conti, Luigi Fortuna, Mattia Frasca, Giuseppina Passiante and Alessandro Rizzo

Durham University, Business School MIll HIll, United Kingdom; University of Catania, Italy; University of Lecce, Italy; Politecnico di Bari, Italy

 

 

 

 

Saturday, July 24,  11:00-12:40

Session S2: Special Session on Spatio-temporal phenomena in PDE based CNN models

11:00-11:20  

Seismic Wave Propagation Modeling on Emulated Digital CNN-UM Architecture

Péter Kozma, Sándor Kocsárdi, Péter Sonkoly and Péter Szolgay

University of Veszprém, Hungary

11:20-11:40  

On the Possibilities of Using Two Grid Coupled CNN's for Face Features Extraction.

Liviu Goras and Paul Ungureanu

"Gh. Asachi" Technical University of Iasi, Romania

11:40-12:00  

On PDE based CNN models with memory 

Angela Slavova

Institute of Mathematics Bulgarian Acad. of Sci., Bulgaria

12:00-12:20  

Linear interplation of nonlinearities in Cellular Neural Networks CNN

Gunter Geis, Michael Reinisch and Ronald Tetzlaff

Institute of Applied Physics, Germany

12:20-12:40   Tactile sensor modeling by using emulated digital CNN-UM 
Zoltán Nagy, Zsófia Szolgay and Péter Szolgay
University of Veszprem, Hungary; Technical University of Budapest, Hungary

 

 

Session A6: Application 6

11:00-11:20  

Translation and Rotation of Grey - Scale Images by means of Analogic Cellular Neural Network

Giovanni Costantini, Daniele Casali, Massimo Carota and Renzo Perfetti

University of Rome "Tor Vergata", Italy; University of Perugia, Italy

11:20-11:40   

Cellular Neural Network Parallelization Rules

Thomas Weishaeupl and Erich Schikuta

University of Vienna, Austria

11:40-12:00   

Inter-Slice Interpolation based on Matching Information, using Cellular Neural Networks

Alexandru Gacsádi, Cristian Grava, Teodor Maghiar, Cornelia Gordan and Adriana Grava

University of Oradea, Romania

12:00-12:20   

CNN as Curve Shortening Flow Computer

István Petrás and Tamás Roska

MTA - Computer and Automation Research Institute, Hungary

12:20-12:40   

FPGA implementation of a DT-CNN-based infrared landmine detection system

Fernando Pardo Seco, Marco Balsi, Paula Lopez Martinez and Diego Cabello Ferrer

University of Santiago de Compostela, Spain; University La Sapienza, Italy; Fraunhofer IIS ICD-A, Germany

 

 

 

 

Saturday, July 24, 14:00-15:40

Session TI: Theory and Implementation

14:00-14:20  

A Floating-gate Approach for Obtaining Accurate Current References in the CNN Cell

Mikko Pankaala, Ari Paasio and Mika Laiho

University of Turku, Finland; University of Turku, Helsinki University of Tech, Finland

14:20-14:40   

TD-DPTA Cellular Neural Network Cell

Fausto Sargeni, Vincenzo Bonaiuto and Maurizio Bonifazi

Dept. of Electronic Engineering - University of, Italy

14:40-15:00   

A Fuzzy-Logic-based Cellular Neural Network Integrated System

Chin-Teng Lin and Chun-Lung Chang

National Chiao-Tung University, Taiwan

15:00-15:20   

Necessary and sufficient conditions for one-dimensional discrete-time binary cellular neural network to be stable

Tetsuo Nishi, Hajime Hara and Norikazu Takahashi

Kyushu University, Japan; Hiroshima Institute of Technology, Japan

15:20-15:40   

Finite Iteration DT-CNN with Stationary Templates

Joerg D. Wichard, Maciej Ogorzalek, Christian Merkwirth and Jochen Broecker

AGH - University of Science and Technologie, Poland; Max-Planck-Institut fuer Informatik, Germany; London School of Economics, United Kingdom

 

 

Session A7: Application 7

14:00-14:20  

Prediction of Brain Electrical Activity in Epilepsy by Multilayer Discrete Time Cellular Neural Networks (DTCNN)

Radu Niederhofer, C. Gollas and Ronald Tetzlaff

Johann Wolfgang Goethe University, Germany

14:20-14:40   

Pattern Recognition in Noisy and Non-stationary Time Series with Cellular Neural Networks

Anton Chernihovskyi, Robert Sowa, Sascha Florin, Christian E. Elger and Klaus Lehnertz

Dept. of Epileptology, University of Bonn, Germany

14:40-15:00   

Deformable Grid Paradigm Implementation in Cellular Neural Networks

Krzysztof Slot and Piotr Korbel

Institute of Electronics, Tech. University of Lodz, Poland

15:00-15:20   

Novel CNN based detection for increased spectral efficiency

János Levendovszky, András Oláh and Domonkos Varga

Pázmány Péter Catholic University, Hungary; Budapest University of Technology and Economics, Hungary

15:20-15:40   

Information flow and self-organization in connected neural networks

Toshihiro Shimizu

Kokushikan University, Japan