Szabolcs Tőkés

Followers

Following

Public Views

Andrej Dujella

University of Zagreb

Hemin Koyi

Uppsala University

Jana Javornik

University of East London

Armando Marques-Guedes

UNL - New University of Lisbon

Graham Martin

University of Leicester

Gwen Robbins Schug

University of North Carolina at Greensboro

Gabriel Gutierrez-Alonso

University of Salamanca

John Sutton

Macquarie University

Eros Carvalho

Universidade Federal do Rio Grande do Sul

Kevin Arbuckle

Swansea University

Uploads

Papers by Szabolcs Tőkés

Media 1: In-line hologram segmentation for volumetric samples

Originally published in Applied Optics on 01 January 2013 (ao-52-1-A45

Computer

Importance and prospects of optical computing and an optical implementation of the CNN-UM

Download

$Research paper thumbnail of Multibeam 2D acousto-optic modulator combined with arrays of diffractive optical elements$

Multibeam 2D acousto-optic modulator combined with arrays of diffractive optical elements

An interference patterning equipment using DOE array

A lézer printer képalkotási hibái és optikai korrekciójuk

On-chip hullámfront érzékelés és processzálás parallel implementációja = Parallel implementation of on-chip wavefront sensing and processing

Download

POAC: optical computer for large data sets

Programozható optoelektronikus tömbprocesszorok (POAC) és alkalmazásaik= Programmable opto-electric array processors (POAC) and their applications

ABSTRACT Kidolgoztunk és megépítettünk egy hordozható programozható opto-elektronikus analogikus ... more ABSTRACT Kidolgoztunk és megépítettünk egy hordozható programozható opto-elektronikus analogikus CNN számítógépet (Laptop-POAC), amelyet céltárgyak felismerésére és követésére alkalmaztunk. A POAC (programmabble optical analogic array computer) magprocesszora egy új típusu optikai korrelátor, amelyben bakteriorhodopsin (BR) filmet alkalmaztunk dinamikus (újraírható, többszörösen olvasható, törölhető) holografikus anyagként. Ez az optikai CNN implementáció egyesíti az optikai számítógépek nagy sebességét, rendkívűl nagy práhuzamosságát (milliós nagyságrendű csatornaszám) és a nagymértű templatek alkalmazhatóságát a CNN eszköszök rugalmas programozhatóságával. Különös jellemzője ennek az optikai tömbprocesszornak az, hogy a programozó templateket vagy egy kétdimenziós akuszto-optikai eltérítővel (ez esetben 64x64 pixel méretű templateket használtunk), vagy egy VCSEL lézer mátrixszal valósíthatjuk meg. A bemenő képeket 600x800 pixel felbontású folyadékkristályos mikromegjelenítővel vittük be. Meghatároztuk a jelenleg beszerezhető kulcselemekkel elérhető maximális felbontás és a sebesség korlátait. Ezáltal kimutatttuk, hogy a kispárhuzamosságú elektronikus adatátvitel (továbbá a létező CCD, CMOS és vizuális CNN chipek felbontás/sebesség) korlátai miatt 2D-s optika utófeldolgozásra van szükség és lehetőség. Új és hatékony céltárgy felismerő és több tárgy kvázi-egyidejű követésére alkalmas algoritmust dolgoztunk ki. Mérésekkel bizonyítottuk a berendezés és az algoritmusok hatékonyságát. | A portable programmable opto-electronic analogic CNN computer (Laptop-POAC) has been built and used to recognize and track targets. Its kernel processor is a novel type of high performance optical correlator based on the use of bacteriorhodopsin (BR) as a dynamic holographic material. This optical CNN implementation combines the optical computer&#39;s high speed, high parallelism (1 000 000 channel) and large applicable template sizes with flexible programmability of the CNN devices. A unique feature of this optical array computer is that programming templates can be applied either by a 2D acousto-optical deflector (up to 64x64 pixel size templates) incoherently or by VCSEL arrays. Input images are fed-in by a LCD-SLM of 600x800 pixel resolution. Evaluation of trade-off between speed and resolution is given. Novel and effective target recognition and multiple-target-tracking algorithms have been developed for the POAC. Tracking experiments are demonstrated. In the present model a video-speed CCD camera is recording the correlograms, however, later a CNN-UM chip and a high-speed CMOS camera will be applied for combined optical/electro-optical post-processing. An optical CNN template- and algorithm-library has been developed to solve a great variety of image processing tasks. It seems to be an important future work to expand this library.

Programmable Analogic Cellular Optical Computer Using Bacteriorhodopsin as Analog Rewriteable lmage Memory

Bacteriorhodopsin has been proved to be an outstanding candidate for reversible, transient, real-... more Bacteriorhodopsin has been proved to be an outstanding candidate for reversible, transient, real-time, holographic material. Here we show our preliminary experimental investigation of its applicability as a transient analog memory, concentrating on its possible utilization in optical CNN implementations and in programmable opto-electronic analogic CNN computers (POAC). Different possible architectures and the technical details of its applicability are also discussed. The main objective is to provide a framework for the implementation of Programmable Opto-Electronic Analogic CNN Computers embedding CNN Universal Chips. Specifically, a new method for optical CNN implementation is provided and some details are experimentally studied. The POAC architecture includes the integration of an optical processing system, such as a joint transform correlator using bacteriorhodopsin as holographic material, with the fast spatio-temporal processing capabilities of a CNN-UM chip. We have built and tested an optical sub-unit of this experimental opto-electronic architecture to examine their processing capabilities in complex target recognition tasks. The main idea is to introduce stored programmability into optical computing. The specification of the necessary holographic material for POAC application is given. Some measurement results on BR samples are presented.

Download

Programmable OASLM as a Novel Sensing Cellular Computer

2006 10th International Workshop on Cellular Neural Networks and Their Applications, 2006

Abstract We have developed a novel type of correlators named searching correlator (SC)(Tokes et a... more Abstract We have developed a novel type of correlators named searching correlator (SC)(Tokes et al., 2003) using bacteriorhodopsin (BR) as a dynamic rewritable and multiply readable holographic material. Our POAC (programmable optical array computer) model can perform highly parallel optical correlations and feedforward CNN operations. However, to keep pace with digital technologies (Moore&amp;amp;#x27;s law); newer, more effective materials are needed. The sensitivity, diffraction efficiency and cycle time are the most critical ...

Multicolor digital holographic microscope (DHM) for biological purposes

Imaging, Manipulation, and Analysis of Biomolecules, Cells, and Tissues VIII, 2010

Our digital holographic (DH) approach can be used to study tissue structures both in vitro and in... more Our digital holographic (DH) approach can be used to study tissue structures both in vitro and in vivo. This DHM architecture can produce three color microscopic 3D and 4D (video) images. We record 3 color (RGB) holograms with single exposures, and the perfect compensation of color crosstalk is solved. An in-line holographic setup and reconstruction algorithms are presented with demonstrative simulations and experimentally captured and numerically reconstructed images. Comparing the individually reconstructed color images with each other can provide information both for recognition of different types of cells or microorganisms, and for diagnostic purposes as well. Experimental example is given observing microscopic hydro-biological organisms using a color digital holographic microscope.

$In the current algorithm however we use the conventional angular spectrum method, which is the solution of the Rayliegh-Sommerfeld scalar diffraction integral.‘ Using E(a,y,0) as the monochromatic electric field distribution at the hologram plane (z = 0) the electric field distribution at a z 4 0 plane will be:$

Download

Optical implementation of the CNN-UM computer and its realization with Joint Fourier Transform Correlator (JTC)

Dennis Gabor as the Initiator of Optical Computing: Importance and Prospects of Optical Computing and an Optical Implementation of the CNN-UM Computer

Design Aspects of an Optical Correlator Based CNN Implementation

Optical correlators provide efficient optical implementation of the CNN computers with high paral... more Optical correlators provide efficient optical implementation of the CNN computers with high parallelism. A semi incoherent optical correlator with design considerations is presented. Theoretical and experimental study of an optical feed-forward processor using bacteriorhodopsin (bR) as a dynamic holographic material and a 2D acousto-optical deflector (AOD) for the implementation of the reconstructing template is given. Special attention is paid for the choice of the optical architecture, what can serve our = = = = goals the best. General definition of the requirements against dynamic holographic materials applied in optical processors is given in a previous paper of us [14] and a detailed specification and functional model of the bacteriorhodopsin based dynamic memory is worked out. Some preliminary correlation results are shown to corroborate the feasibility of this approach.

Download

A Study of Correlation Process of a Modified JTC for Optical Cellular Neural Networks

ahmedayoub.webs.com

Abstract–Optical computing, especially calculations based on modified Joint Transform Correlators... more Abstract–Optical computing, especially calculations based on modified Joint Transform Correlators (t2-JTC) where the t2 template is applied in the recognition step, provides a real time and powerful processing module to enhance the Cellular Neural Networks (CNN)[1] parallelism. In this work, we addressed key building blocks of JTC (correlation process, optimal postprocessing module and thresholding techniques) via optical simulations. The results reflect the possibility of implementing an optical CNN with several image ...

Download

An advanced joint Fourier transform correlator (JTC)

DIFFRACTIVE OPTICS 2001, 2001

Purpose of this paper to present an improved JTC architecture. This architecture combines advanta... more Purpose of this paper to present an improved JTC architecture. This architecture combines advantages of the incoherent and coherent optical correlators meanwhile some of their drawbacks are eliminated. Our ultimate purpose is to use this special correlator to efficiently implement optically a feedforward-only CNN [1]. Otherwise, CNN paradigm [2, 3] provides efficient algorithmic frame for optical correlators. First we introduce our new optical correlator achitecture. We shall describe shortly the basic characteristics of the applied equipments ...

Download

1Programmable Opto-electronic CNN implementation provides a new and powerful tool for image processing applications

Abstract. An opto-electronic CNN implementation, as it combines the huge inher-ent computational ... more Abstract. An opto-electronic CNN implementation, as it combines the huge inher-ent computational capabilities of a massively parallel optical system with pro-grammability of the CNN-UM systems, provides an efficient frame for diverse im-age processing applications. Our optical CNN implementation is based on a new, semi incoherent optical correlator architecture that is superior from several points of view to any alternative optical correlator. This architecture provides the possi-bility of real time reprogramming of the correlator like in the joint Fourier trans-form correlators, but preserves the speed of VanderLugt type of systems. Pro-gramming is done through fast holographic read-out by B-templates. With combi-nation of a high speed CNN-UM sensor chip it forms a programmable opto-electronic analogic computer (POAC) architecture. This device is especially useful in those image-processing algorithms, which requires a great number of pattern matching tasks. 1

Download

Design Aspects of an Optical Correlator Based CNN

Programmable opto-electronic CNN implementation provides a new and powerful tool for image processing applications. (Research report of the Analogical and Neural Computing Laboratory DNS-9-2001.)

An opto-electronic CNN implementation, as it combines the huge inherent computational capabilitie... more An opto-electronic CNN implementation, as it combines the huge inherent computational capabilities of a massively parallel optical system with programmability of the CNN-UM systems, provides an efficient frame for diverse image processing applications. Our optical CNN implementation is based on a new, semi incoherent optical correlator architecture that is superior from several points of view to any alternative optical correlator. This architecture provides the possibility of real time reprogramming of the correlator like in the joint Fourier transform correlators, but preserves the speed of VanderLugt type of systems. Programming is done through fast holographic read-out by B-templates. With combination of a high speed CNN-UM sensor chip it forms a programmable optoelectronic analogic computer (POAC) architecture. This device is especially useful in those image-processing algorithms, which requires a great number of pattern matching tasks.

Download

An advanced joint Fourier transform correlator (JTC)

Proceedings of DO, 2001

Media 1: In-line hologram segmentation for volumetric samples

Originally published in Applied Optics on 01 January 2013 (ao-52-1-A45

Computer

Importance and prospects of optical computing and an optical implementation of the CNN-UM

Download

$Research paper thumbnail of Multibeam 2D acousto-optic modulator combined with arrays of diffractive optical elements$

Multibeam 2D acousto-optic modulator combined with arrays of diffractive optical elements

An interference patterning equipment using DOE array

A lézer printer képalkotási hibái és optikai korrekciójuk

On-chip hullámfront érzékelés és processzálás parallel implementációja = Parallel implementation of on-chip wavefront sensing and processing

Download

POAC: optical computer for large data sets

Programozható optoelektronikus tömbprocesszorok (POAC) és alkalmazásaik= Programmable opto-electric array processors (POAC) and their applications

Programmable Analogic Cellular Optical Computer Using Bacteriorhodopsin as Analog Rewriteable lmage Memory

Download

Programmable OASLM as a Novel Sensing Cellular Computer

2006 10th International Workshop on Cellular Neural Networks and Their Applications, 2006

Multicolor digital holographic microscope (DHM) for biological purposes

Imaging, Manipulation, and Analysis of Biomolecules, Cells, and Tissues VIII, 2010

Download

Optical implementation of the CNN-UM computer and its realization with Joint Fourier Transform Correlator (JTC)

Dennis Gabor as the Initiator of Optical Computing: Importance and Prospects of Optical Computing and an Optical Implementation of the CNN-UM Computer

Design Aspects of an Optical Correlator Based CNN Implementation

Download

A Study of Correlation Process of a Modified JTC for Optical Cellular Neural Networks

ahmedayoub.webs.com

Download

An advanced joint Fourier transform correlator (JTC)

DIFFRACTIVE OPTICS 2001, 2001

Download

1Programmable Opto-electronic CNN implementation provides a new and powerful tool for image processing applications

Download

Design Aspects of an Optical Correlator Based CNN

Programmable opto-electronic CNN implementation provides a new and powerful tool for image processing applications. (Research report of the Analogical and Neural Computing Laboratory DNS-9-2001.)