Pub Date : 2002-07-22DOI: 10.1109/CNNA.2002.1035108
G. Liñán, Á. Rodríguez-Vázquez, S. Espejo, R. Domínguez-Castro, E. Roca
This paper presents a new multimode optical sensor architecture for the optical interface of Visual CNN (cellular neural net) chips. The sensor offers to the user the possibility of choosing the photo-sensitive device as well as the mechanism for transducing the photogenerated charges into the correspondent pixel voltage. Both linear or logarithmic compression acquisition modes are available. This makes the sensor very suitable to be used in very different illumination conditions.
{"title":"A multimode gray-scale CMOS optical sensor for visual computers","authors":"G. Liñán, Á. Rodríguez-Vázquez, S. Espejo, R. Domínguez-Castro, E. Roca","doi":"10.1109/CNNA.2002.1035108","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035108","url":null,"abstract":"This paper presents a new multimode optical sensor architecture for the optical interface of Visual CNN (cellular neural net) chips. The sensor offers to the user the possibility of choosing the photo-sensitive device as well as the mechanism for transducing the photogenerated charges into the correspondent pixel voltage. Both linear or logarithmic compression acquisition modes are available. This makes the sensor very suitable to be used in very different illumination conditions.","PeriodicalId":387716,"journal":{"name":"Proceedings of the 2002 7th IEEE International Workshop on Cellular Neural Networks and Their Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131641667","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-22DOI: 10.1109/CNNA.2002.1035109
M. Salerno, F. Sargeni, V. Bonaiuto
The implementation of an artificial vision algorithm in real time is really attractive in such an application as the field of environment sensing. The SVCNN (stereo vision cellular neural network) chip is an analogue circuit able to compute in real time the Disparity Map from a couple of images by using a stereo visual system algorithm. A "test-bed" board for the 16/spl times/64 SVCNN chip is presented in this paper. This board is composed of an analogue processing core implemented by two 16/spl times/64 SVCNN chips together with a digital high performance pre-processing unit and a video grabbing section.
{"title":"Test-bed board for 16/spl times/64 stereo vision CNN chip","authors":"M. Salerno, F. Sargeni, V. Bonaiuto","doi":"10.1109/CNNA.2002.1035109","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035109","url":null,"abstract":"The implementation of an artificial vision algorithm in real time is really attractive in such an application as the field of environment sensing. The SVCNN (stereo vision cellular neural network) chip is an analogue circuit able to compute in real time the Disparity Map from a couple of images by using a stereo visual system algorithm. A \"test-bed\" board for the 16/spl times/64 SVCNN chip is presented in this paper. This board is composed of an analogue processing core implemented by two 16/spl times/64 SVCNN chips together with a digital high performance pre-processing unit and a video grabbing section.","PeriodicalId":387716,"journal":{"name":"Proceedings of the 2002 7th IEEE International Workshop on Cellular Neural Networks and Their Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128758224","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-22DOI: 10.1109/CNNA.2002.1035040
L. Goras, R. Ghinea, T. Dinu, T. David, E. David
The stability and dynamics of a class of cellular neural networks (CNN) in the central linear part is investigated using the decoupling technique based on discrete spatial transforms, Nyquist and root locus techniques.
{"title":"On the dynamics of a class of cellular neural networks","authors":"L. Goras, R. Ghinea, T. Dinu, T. David, E. David","doi":"10.1109/CNNA.2002.1035040","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035040","url":null,"abstract":"The stability and dynamics of a class of cellular neural networks (CNN) in the central linear part is investigated using the decoupling technique based on discrete spatial transforms, Nyquist and root locus techniques.","PeriodicalId":387716,"journal":{"name":"Proceedings of the 2002 7th IEEE International Workshop on Cellular Neural Networks and Their Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115740147","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-22DOI: 10.1109/CNNA.2002.1035077
G. Tímár, K. Karacs, C. Rekeczky
This report describes analogic algorithms used in the preprocessing and segmentation phase of offline handwriting recognition tasks. The handwriting recognition approach is segmentation based, i.e. it attempts to segment words into their constituent letters. In order to improve their speed the utilized CNN algorithms use dynamic, wave front propagation-based methods instead of relying on morphologic operators embedded into iterative algorithms. The system first locates handwritten lines in the page image then corrects their skew as necessary. Afterwards it searches for words within the lines and corrects skew at the word level as well. A novel trigger wave-based word segmentation algorithm is presented which operates on the skeletons of words. Sample results of experiments conducted on a database of 25 handwritten pages are presented.
{"title":"Analogic preprocessing and segmentation algorithms for off-line handwriting recognition","authors":"G. Tímár, K. Karacs, C. Rekeczky","doi":"10.1109/CNNA.2002.1035077","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035077","url":null,"abstract":"This report describes analogic algorithms used in the preprocessing and segmentation phase of offline handwriting recognition tasks. The handwriting recognition approach is segmentation based, i.e. it attempts to segment words into their constituent letters. In order to improve their speed the utilized CNN algorithms use dynamic, wave front propagation-based methods instead of relying on morphologic operators embedded into iterative algorithms. The system first locates handwritten lines in the page image then corrects their skew as necessary. Afterwards it searches for words within the lines and corrects skew at the word level as well. A novel trigger wave-based word segmentation algorithm is presented which operates on the skeletons of words. Sample results of experiments conducted on a database of 25 handwritten pages are presented.","PeriodicalId":387716,"journal":{"name":"Proceedings of the 2002 7th IEEE International Workshop on Cellular Neural Networks and Their Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114606784","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-22DOI: 10.1109/CNNA.2002.1035049
Z. Nagy, P. Szolgay
A new emulated digital multi-layer CNN-UM chip architecture called Falcon has been developed. In this paper the main steps of the FPGA implementation are introduced. Main results are as follows: CNN-UM architecture emulated on Xilinx Virtex series FPGA, 3D non-linear spatio-temporal dynamics can be implemented on this architecture. The critical parameters of the implementation in single layer configuration are 55 million cell update/second/processor core or equivalently 1 GOPS computing performance. In face of the high performance the power requirements of the architecture are relatively low only /spl sim/3 W per processor core. Using re-configurable devices to implement emulated digital architectures provides more flexibility compared to the custom VLSI designs because different Falcon architectures can be used on the same FPGA device.
{"title":"Configurable multi-layer CNN-UM emulator on FPGA","authors":"Z. Nagy, P. Szolgay","doi":"10.1109/CNNA.2002.1035049","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035049","url":null,"abstract":"A new emulated digital multi-layer CNN-UM chip architecture called Falcon has been developed. In this paper the main steps of the FPGA implementation are introduced. Main results are as follows: CNN-UM architecture emulated on Xilinx Virtex series FPGA, 3D non-linear spatio-temporal dynamics can be implemented on this architecture. The critical parameters of the implementation in single layer configuration are 55 million cell update/second/processor core or equivalently 1 GOPS computing performance. In face of the high performance the power requirements of the architecture are relatively low only /spl sim/3 W per processor core. Using re-configurable devices to implement emulated digital architectures provides more flexibility compared to the custom VLSI designs because different Falcon architectures can be used on the same FPGA device.","PeriodicalId":387716,"journal":{"name":"Proceedings of the 2002 7th IEEE International Workshop on Cellular Neural Networks and Their Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114826160","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-22DOI: 10.1109/CNNA.2002.1035033
M. Di Marco, M. Forti, A. Tesi
It is known that symmetric cellular neural networks (CNNs) are completely stable, i.e., each trajectory converges towards some equilibrium point. The paper addresses the issue of the loss of CNN complete stability caused by errors in the implementation of the nominal symmetric interconnections. The main result is a structural condition which implies the existence of stable limit cycles generated via Hopf bifurcations, even for arbitrarily small perturbations of the nominal interconnections. Furthermore, analytic results providing an approximate relationship between the limit cycle features and the fundamental CNN parameters are presented.
{"title":"A study on limit cycles in nearly symmetric cellular neural networks","authors":"M. Di Marco, M. Forti, A. Tesi","doi":"10.1109/CNNA.2002.1035033","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035033","url":null,"abstract":"It is known that symmetric cellular neural networks (CNNs) are completely stable, i.e., each trajectory converges towards some equilibrium point. The paper addresses the issue of the loss of CNN complete stability caused by errors in the implementation of the nominal symmetric interconnections. The main result is a structural condition which implies the existence of stable limit cycles generated via Hopf bifurcations, even for arbitrarily small perturbations of the nominal interconnections. Furthermore, analytic results providing an approximate relationship between the limit cycle features and the fundamental CNN parameters are presented.","PeriodicalId":387716,"journal":{"name":"Proceedings of the 2002 7th IEEE International Workshop on Cellular Neural Networks and Their Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126763372","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-22DOI: 10.1109/CNNA.2002.1035060
R. Dogaru, I. Dogaru
This paper investigates a novel application in biometry of the emergent complex patterns in cellular neural networks. Using a set of patterns generated in the "unstable" regime of discrete-time generalized cellular automata (GCA), it was found that human subjects exposed to these patterns would resonate perceptually with them in a personal manner which qualifies such a resonance as a biometric "brain signature" of a person. Further research indicates that by using an large enough set of stimuli, a simple linear perceptron can discriminate among persons, thus performing authentication at a low implementation cost.
{"title":"Biometric authentication based on perceptual resonance between CNN emergent patterns and humans","authors":"R. Dogaru, I. Dogaru","doi":"10.1109/CNNA.2002.1035060","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035060","url":null,"abstract":"This paper investigates a novel application in biometry of the emergent complex patterns in cellular neural networks. Using a set of patterns generated in the \"unstable\" regime of discrete-time generalized cellular automata (GCA), it was found that human subjects exposed to these patterns would resonate perceptually with them in a personal manner which qualifies such a resonance as a biometric \"brain signature\" of a person. Further research indicates that by using an large enough set of stimuli, a simple linear perceptron can discriminate among persons, thus performing authentication at a low implementation cost.","PeriodicalId":387716,"journal":{"name":"Proceedings of the 2002 7th IEEE International Workshop on Cellular Neural Networks and Their Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127829539","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-22DOI: 10.1109/CNNA.2002.1035066
G. Cserey, C. Rekeczky, P. Foldesy
This paper describes parallel histogram modification techniques with embedded morphological preprocessing methods within the CNN-UM framework. The procedure is formulated in terms of nonlinear partial differential equations (PDE) and approximated through finite differences in space, resulting in coupled nonlinear ordinary differential equations (ODE). The I/O mapping of the system (containing both local and global couplings) can be calculated by a complex analogic (analog and logic) algorithm executed on a stored program nonlinear array processor, called the cellular nonlinear network universal machine (CNN-UM). We describe and illustrate how implementation of the algorithm results in an adaptive multi-thresholding scheme when histogram modification is combined with embedded morphological processing at a finite (low) number of grayscale levels. This has obvious advantages if the further processing steps are segmentation and/or recognition. Experimental results processing real-life and echocardiography images are measured on different hardware/software platforms, including a 64/spl times/64 CNN-UM chip (ACE4k).
{"title":"PDE based histogram modification with embedded morphological processing of the level-sets","authors":"G. Cserey, C. Rekeczky, P. Foldesy","doi":"10.1109/CNNA.2002.1035066","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035066","url":null,"abstract":"This paper describes parallel histogram modification techniques with embedded morphological preprocessing methods within the CNN-UM framework. The procedure is formulated in terms of nonlinear partial differential equations (PDE) and approximated through finite differences in space, resulting in coupled nonlinear ordinary differential equations (ODE). The I/O mapping of the system (containing both local and global couplings) can be calculated by a complex analogic (analog and logic) algorithm executed on a stored program nonlinear array processor, called the cellular nonlinear network universal machine (CNN-UM). We describe and illustrate how implementation of the algorithm results in an adaptive multi-thresholding scheme when histogram modification is combined with embedded morphological processing at a finite (low) number of grayscale levels. This has obvious advantages if the further processing steps are segmentation and/or recognition. Experimental results processing real-life and echocardiography images are measured on different hardware/software platforms, including a 64/spl times/64 CNN-UM chip (ACE4k).","PeriodicalId":387716,"journal":{"name":"Proceedings of the 2002 7th IEEE International Workshop on Cellular Neural Networks and Their Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127246209","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-22DOI: 10.1109/CNNA.2002.1035056
Christian E. Elger, F. Mormann, T. Kreuz, R. Andrzejak, Christoph Rieke, Robert Sowa, S. Florin, P. David, K. Lehnertz
In this overview we present our work investigating the spatio-temporal dynamics of the epileptogenic process using time series analysis techniques derived from the theory of nonlinear dynamics. Apart from a localization of epileptic foci in different anatomical locations during the seizure-free interval we discuss possibilities for seizure prediction, a field that represents one of the greatest challenges in epileptology. The unequivocal definition of a pre-seizure state of a sufficient length would enable investigation of basic mechanisms leading to seizure initiation in humans and provide a basis for the development of adequate seizure prevention strategies.
{"title":"Characterizing the spatio-temporal dynamics of the epileptogenic process with nonlinear EEG analyses","authors":"Christian E. Elger, F. Mormann, T. Kreuz, R. Andrzejak, Christoph Rieke, Robert Sowa, S. Florin, P. David, K. Lehnertz","doi":"10.1109/CNNA.2002.1035056","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035056","url":null,"abstract":"In this overview we present our work investigating the spatio-temporal dynamics of the epileptogenic process using time series analysis techniques derived from the theory of nonlinear dynamics. Apart from a localization of epileptic foci in different anatomical locations during the seizure-free interval we discuss possibilities for seizure prediction, a field that represents one of the greatest challenges in epileptology. The unequivocal definition of a pre-seizure state of a sufficient length would enable investigation of basic mechanisms leading to seizure initiation in humans and provide a basis for the development of adequate seizure prevention strategies.","PeriodicalId":387716,"journal":{"name":"Proceedings of the 2002 7th IEEE International Workshop on Cellular Neural Networks and Their Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127357920","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-22DOI: 10.1109/CNNA.2002.1035095
P. Kozma, B. Kránicz, P. Szolgay
Nowadays many problems requiring huge computing power have risen. Although the performance of digital processors doubles every year, there are certain tasks where the computation cannot be carried out within a reasonable time interval. Such hard problems are the analysis of big dynamical systems or real-time exact colour reproduction. The exact colour visualization of motion pictures is necessary in industrial, medical and scientific research areas. Thus, for example, exact colour reproduction is required for remote medical diagnosis or remote operation. The doctor has to see the same image that appears in reality. Device dependent colour appearance may cause faulty decisions. Nowadays these problems cannot be solved perfectly because many steps of the transformation are not completely known and the huge number of computations cannot be done in real-time even by the fastest PC. In this article we describe some methods to produce exact colours in a remote medical diagnostic system.
{"title":"Colour space transformation and exact colour reproduction with CNN technology","authors":"P. Kozma, B. Kránicz, P. Szolgay","doi":"10.1109/CNNA.2002.1035095","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035095","url":null,"abstract":"Nowadays many problems requiring huge computing power have risen. Although the performance of digital processors doubles every year, there are certain tasks where the computation cannot be carried out within a reasonable time interval. Such hard problems are the analysis of big dynamical systems or real-time exact colour reproduction. The exact colour visualization of motion pictures is necessary in industrial, medical and scientific research areas. Thus, for example, exact colour reproduction is required for remote medical diagnosis or remote operation. The doctor has to see the same image that appears in reality. Device dependent colour appearance may cause faulty decisions. Nowadays these problems cannot be solved perfectly because many steps of the transformation are not completely known and the huge number of computations cannot be done in real-time even by the fastest PC. In this article we describe some methods to produce exact colours in a remote medical diagnostic system.","PeriodicalId":387716,"journal":{"name":"Proceedings of the 2002 7th IEEE International Workshop on Cellular Neural Networks and Their Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126376701","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}