- István Szatmári, Ph.D., senior research fellow
István Szatmári received the M.Sc. degree in electrical engineering from the Technical University of Budapest, Hungary, in 1995 and the Ph.D. degree from the Hungarian Academy of Sciences in 2002. He is currently at the Analogical and Neural Computing Laboratory of the Computer and Automation Research Institute of the Hungarian Academy of Sciences. In 1997 and 2004, he worked at the University of California (Berkeley, USA) as a research assistant working on CNN projects related to real time image processing. His research interests include associative memories, theory of metrics, and hardware design hosting CNN-UM chips.
He is the author or co-author of more than 30 publications, out of these 8 were published in referred journals and 15 in conference proceedings. He is a co-author of the "Towards the Visual Microprocessor: VLSI Design and the Use of Cellular Neural Network Universal Machines" book from Wiley.
He received Bolyai János Hungarian State Fellowship (2000), and Hungarian State Eötvös Scholarship (1997). He is member of the IEEE.
In recent years he has participated in several successful research and development projects including US, European, Hungarian R&D grants. His development experience and competency includes mix mode (analog and digital) architecture design and measurement, multiprocessor Texas DSP environment, fast image processing algorithm development.
Publications[ order by time]
[ order by categories ]
[ order by authors]
Book chapter- A CNN application development environment and toolkit, CADETWin.
Authors: Roska, T; László, K; Kék, L; Kozek, T; Nemes, L; Rekeczky, C; Szatmári, I; Zarándy, A; Zöld, S; Szolgay, PDepartment: Cellular Sensory and Optical Wave Computing LaboratoryDate: 2000.Published by: Towards the visual microprocessor. VLSI design and the use of cellular neural network universal machines. (Page: 39-58)
Conference issue- Pattern formation in oscillatory media: beyond reaction-diffusion model.
- GPU implementation of volume reconstruction and object detection in digital holographic microscopy.
- High-speed label inspection system for textile industry
- Efficient off-line feature selection strategies for on-line classifier systems
- Topographic and non-topographic neural network based computational platform for UAV applications
- Active wave computing on silicon: chip experiments
- Adaptive multi-rate, multi-grid and multi-scale algorithms running on analogic architecture
- Classification of spatio-temporal features: the nearest neighbor family
- Feature guided visual attention with topographic array processing and neural network-based classification
- Vision systems based on the 128x128 focal plane cellular visual microprocessor chips
- Multi-channel spatio-temporal topographic processing for visual search and navigation
- Bio-inspired flight control and visual search with CNN technology
- Image processing library for the ALADDIN visual computer.
- Analogic cellular PDE machines.
- Moving object traking on panoramic images.
- Computing on silicon with trigger-waves: experiments on CNN-UM chips.
- The implementation of a nonlinear wave metric for image analysis and classification on the 64x64 I/O CNN-UM chip.
- CNN technology in action.
Authors: Zarándy, Á; Espejo, S; Földesy, P; Kék, L; Linán, G; Rekeczky, C; Rodriguez-Vázquez, A; Roska, T; Szatmári, I; Szirányi, T; Szolgay, PDepartment: Cellular Sensory and Optical Wave Computing LaboratoryDate: 2000.Published by: Proceedings of the 6th IEEE international workshop on cellular neural networks and their applications. (CNNA 2000). Catania, 2000. (Page: 79-81)
- An analogic CNN engine board with the 64x64 analog I/O CNN-UM chip.
- The CADETWin application software design system - a tutorial.
Authors: Szolgay, P; László, K; Kék, L; Kozek, T; Nemes, L; Petrás, I; Rekeczky, CS; Szatmári, I; Zarándy, Á; Zöld, S; Roska, TDepartment: Cellular Sensory and Optical Wave Computing LaboratoryDate: 1999.Published by: Design automation day on cellular visual microprocessor. Stresa, 1999. (Page: 58-68)
- The computational infrastructure for cellular visual microprocessors.
Authors: Szolgay, P; Zarándy, Á; Zöld, S; Roska, T; Földesy, P; Kék, L; Kozek, T; László, K; Petrás, I; Rekeczky, CS; Szatmári, I; Bálya, DDepartment: Cellular Sensory and Optical Wave Computing LaboratoryDate: 1999.Published by: MicroNeuro '99. Proceedings of the seventh international conference on microelectronics for neural, fuzzy, and bio-inspired systems. Granada, 1999. (Page: 54-60)
- Spatio-temporal CNN algorithm for object segmentation and object recognition.
- The CNN implementation of wave type metric for image analysis and classification.
- Image segmentation and edge detection via constrained diffusion and adaptive morphology: a CNN approach to bubble/debris image enhancement.
- Bubble-debris classification via binary morphology and autowave metric on CNN.
- ACE: A digital floating point CNN emulator engine.
Authors: Fehér, B; Szolgay, P; Roska, T; Radványi, AG; Szirányi, T; Csapodi, M; László, K; Nemes, L; Szatmári, I; Tóth, G; Venetianer, PLDepartment: Cellular Sensory and Optical Wave Computing LaboratoryDate: 1996.Published by: 1996 fourth IEEE international workshop on Cellular Neural Networks and their applications proceedings. CNNA-96. Seville, 1996. (Page: 273-278)
- A fast learning method to implement associative memory on CNN
New search
|
Search 
Magyar
