Pub Date : 2002-07-22DOI: 10.1109/CNNA.2002.1035086
K. Mainzer
Cellular neural/nonlinear networks (CNN) are considered as the emergence of a new paradigm of complexity in the information age. In the framework of nonlinear dynamical systems, they can be compared with other paradigms of complexity (e.g., synergetics) as far as they are mathematically formalized. The dogma of local activity demonstrates remarkable advantages for computer simulations of pattern formation and pattern recognition in nature (e.g., diffusion-reaction processes, morphogenesis, artificial life, neural networks), but especially for nonlinear information processing in computer and chip technology. In the information age, nonlinear information processing and communication in global networks like the Internet are a challenge of complexity management. Ubiquitous computing is the future of a globalized world. The recent debate in sociology, economics, and philosophy on 'globalism' and 'localism' underlines the importance of the CNN paradigm for social, economic, and cultural systems.
{"title":"CNN and the evolution of complex information systems in nature and technology","authors":"K. Mainzer","doi":"10.1109/CNNA.2002.1035086","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035086","url":null,"abstract":"Cellular neural/nonlinear networks (CNN) are considered as the emergence of a new paradigm of complexity in the information age. In the framework of nonlinear dynamical systems, they can be compared with other paradigms of complexity (e.g., synergetics) as far as they are mathematically formalized. The dogma of local activity demonstrates remarkable advantages for computer simulations of pattern formation and pattern recognition in nature (e.g., diffusion-reaction processes, morphogenesis, artificial life, neural networks), but especially for nonlinear information processing in computer and chip technology. In the information age, nonlinear information processing and communication in global networks like the Internet are a challenge of complexity management. Ubiquitous computing is the future of a globalized world. The recent debate in sociology, economics, and philosophy on 'globalism' and 'localism' underlines the importance of the CNN paradigm for social, economic, and cultural systems.","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":"125112037","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.2002.1035031
I. Petrás, T. Roska, L. O. Chua
In this paper we introduce a new experimental tool, a real-time programmable, spatial-temporal bifurcation test-bed. We present the experimental analysis of an antisymmetric template class. This class produces novel spatial-temporal patterns that have complex dynamics. The character of these propagating patterns depends on the self-feedback and on the sign of the coupling below the self-feedback template element. We also show how to use these patterns for morphological detection.
{"title":"New spatial-temporal patterns and the first programmable on-chip bifurcation test-bed","authors":"I. Petrás, T. Roska, L. O. Chua","doi":"10.1109/CNNA.2002.1035031","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035031","url":null,"abstract":"In this paper we introduce a new experimental tool, a real-time programmable, spatial-temporal bifurcation test-bed. We present the experimental analysis of an antisymmetric template class. This class produces novel spatial-temporal patterns that have complex dynamics. The character of these propagating patterns depends on the self-feedback and on the sign of the coupling below the self-feedback template element. We also show how to use these patterns for morphological detection.","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":"130602147","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.2002.1035068
Wen-Cheng Yen, Rongna Chen, Jui-Lin Lai
The first VLSI implementation of the fuzzy cellular neural network (FCNN) structure is presented. The MIN/MAX CNN (MMCNN) is a special case of type-II FCNN, which consists only of local MIN and MAX operations. Due to the simple structure of the MMCNN, it is very suitable for VLSI implementation in image processing. Only one neuron cell, two multipliers, and nine min/max circuits realize the proposed MMCNN. Correct functions of the MMCNN in the erosion and dilation of the gray-scale mathematical morphology operation have been successfully verified in HSPICE simulation. FCNNs have great potential in the VLSI implementation of neural network systems in various signal processing applications.
{"title":"Design of MIN/MAX cellular neural networks (MMCNNS) in CMOS technology","authors":"Wen-Cheng Yen, Rongna Chen, Jui-Lin Lai","doi":"10.1109/CNNA.2002.1035068","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035068","url":null,"abstract":"The first VLSI implementation of the fuzzy cellular neural network (FCNN) structure is presented. The MIN/MAX CNN (MMCNN) is a special case of type-II FCNN, which consists only of local MIN and MAX operations. Due to the simple structure of the MMCNN, it is very suitable for VLSI implementation in image processing. Only one neuron cell, two multipliers, and nine min/max circuits realize the proposed MMCNN. Correct functions of the MMCNN in the erosion and dilation of the gray-scale mathematical morphology operation have been successfully verified in HSPICE simulation. FCNNs have great potential in the VLSI implementation of neural network systems in various signal processing applications.","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":"134158238","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.2002.1035102
Chiu-Hung Cheng, Chung-Yu Wu
In this paper, a new type of the ratio-memory cellular neural network (RMCNN) with spatial-dependent self-feedback A-template weights is proposed and designed to recognize and classify the black-white image patterns. In the proposed RMCNN, the combined four-quadrant multiplier and two-quadrant divider with separated magnitude and sign is used to implement the Hebbian learning function and the ratio memory. To enhance the capability of pattern learning and recognition from noisy input patterns, the Z-template and the spatial-dependent self-feedback weights in the template A are applied to the proposed new type of RMCNN. The pattern learning and recognition function of the 18/spl times/18 RMCNN is simulated by Matlab software. It has been verified that the advanced RMCNN has the advantages of more stored patterns for recognition, and better recovery rate as compared to the original RMCNN. Thus the proposed RMCNN has great potential in the applications of neural associate memory for image processing.
{"title":"The design of ratio-memory cellular neural network (RMCNN) with self-feedback template weight for pattern learning and recognition","authors":"Chiu-Hung Cheng, Chung-Yu Wu","doi":"10.1109/CNNA.2002.1035102","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035102","url":null,"abstract":"In this paper, a new type of the ratio-memory cellular neural network (RMCNN) with spatial-dependent self-feedback A-template weights is proposed and designed to recognize and classify the black-white image patterns. In the proposed RMCNN, the combined four-quadrant multiplier and two-quadrant divider with separated magnitude and sign is used to implement the Hebbian learning function and the ratio memory. To enhance the capability of pattern learning and recognition from noisy input patterns, the Z-template and the spatial-dependent self-feedback weights in the template A are applied to the proposed new type of RMCNN. The pattern learning and recognition function of the 18/spl times/18 RMCNN is simulated by Matlab software. It has been verified that the advanced RMCNN has the advantages of more stored patterns for recognition, and better recovery rate as compared to the original RMCNN. Thus the proposed RMCNN has great potential in the applications of neural associate memory for image processing.","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":"134051803","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.2002.1035073
E. Saatci, V. Tavsanoglu
Fingerprint images are usually worsened by various kinds of noise causing cracks, scratches and bridges in the ridges as well as ink blurs. These cause matching errors in fingerprint recognition. For effective recognition the correct ridge pattern is essential, requiring the enhancement of fingerprint images. A fingerprint pattern consists of ridges. Segment by segment analysis of the pattern yields various ridge directions and frequencies. By selecting a directional filter with correct filter parameters to match ridge features at each point, we can effectively enhance fingerprint ridges. This paper proposes fingerprint image enhancement based on CNN Gabor-type filters.
{"title":"Fingerprint image enhancement using CNN Gabor-Type filters","authors":"E. Saatci, V. Tavsanoglu","doi":"10.1109/CNNA.2002.1035073","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035073","url":null,"abstract":"Fingerprint images are usually worsened by various kinds of noise causing cracks, scratches and bridges in the ridges as well as ink blurs. These cause matching errors in fingerprint recognition. For effective recognition the correct ridge pattern is essential, requiring the enhancement of fingerprint images. A fingerprint pattern consists of ridges. Segment by segment analysis of the pattern yields various ridge directions and frequencies. By selecting a directional filter with correct filter parameters to match ridge features at each point, we can effectively enhance fingerprint ridges. This paper proposes fingerprint image enhancement based on CNN Gabor-type filters.","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":"132190536","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.2002.1035080
L. Carnimeo, A. Giaquinto
In this paper a cellular fuzzy associative memory containing fuzzy rules for gray image fuzzification in automatic vision systems is developed. This cellular processor is viewed as a subsystem of a CNN-based architecture, which aims to store both bidimensional patterns and the rules to process them. After establishing the fuzzy rules which define the fuzzy associative memory for image processing, a CNN behaving as a memory is synthesized to store them. A numerical example is reported to show how the synthesized cellular FAM can process bidimensional patterns for robotic navigation purposes.
{"title":"A cellular fuzzy associative memory for bidimensional pattern segmentation","authors":"L. Carnimeo, A. Giaquinto","doi":"10.1109/CNNA.2002.1035080","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035080","url":null,"abstract":"In this paper a cellular fuzzy associative memory containing fuzzy rules for gray image fuzzification in automatic vision systems is developed. This cellular processor is viewed as a subsystem of a CNN-based architecture, which aims to store both bidimensional patterns and the rules to process them. After establishing the fuzzy rules which define the fuzzy associative memory for image processing, a CNN behaving as a memory is synthesized to store them. A numerical example is reported to show how the synthesized cellular FAM can process bidimensional patterns for robotic navigation purposes.","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":"133382755","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.2002.1035094
T. Szirányi
CNN's fast image processing technology helps us to run high-speed filtering tasks for image enhancement, recognition or segmentation. Texture analysis is a specific task, since the whole image is processed massively parallel while we have a limited number of texture-specific filtering and evaluation steps. Former results of simulations and recognition results of simple CNN chips show that the CNN is an appropriate tool for this image-processing task. Now we see what the gray-scale image processor CNN chip at its limited memory capability and data-handling/-processing accuracy can complete for multi-texture images. We demonstrate and compare some of our earlier CNN-related texture analysis methods. Some methods to improve CNN configuration are proposed.
{"title":"Texture segmentation by the 64/spl times/64 CNN chip","authors":"T. Szirányi","doi":"10.1109/CNNA.2002.1035094","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035094","url":null,"abstract":"CNN's fast image processing technology helps us to run high-speed filtering tasks for image enhancement, recognition or segmentation. Texture analysis is a specific task, since the whole image is processed massively parallel while we have a limited number of texture-specific filtering and evaluation steps. Former results of simulations and recognition results of simple CNN chips show that the CNN is an appropriate tool for this image-processing task. Now we see what the gray-scale image processor CNN chip at its limited memory capability and data-handling/-processing accuracy can complete for multi-texture images. We demonstrate and compare some of our earlier CNN-related texture analysis methods. Some methods to improve CNN configuration are proposed.","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":"131771157","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.2002.1035089
M. Zaghloul
This paper reviews recent microfabrication techniques with applications to MEMS, and microsystems. Examples of MEMS devices are presented and their applications to RF-communications are discussed, along with new emerging technologies and their impact on future devices and future architectures.
{"title":"MEMS, microsystems and nanosystems","authors":"M. Zaghloul","doi":"10.1109/CNNA.2002.1035089","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035089","url":null,"abstract":"This paper reviews recent microfabrication techniques with applications to MEMS, and microsystems. Examples of MEMS devices are presented and their applications to RF-communications are discussed, along with new emerging technologies and their impact on future devices and future architectures.","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":"117293230","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.2002.1035072
L. Fortuna, D. Porto
In this paper we consider coupled quantum-dot cells, which are usually used for quantum-dot cellular automata (QCA), as a build unit to construct an analog cellular neural network. It is also shown how simple connection of few quantum-dot cells (even two of them) can cause the onset of chaotic oscillation only with small differences of polarizations and template between cells. An example of polarization evolution in two adjacent cells is reported together with proof of their chaotic behavior.
{"title":"Chaotic phenomena in quantum cellular neural networks","authors":"L. Fortuna, D. Porto","doi":"10.1109/CNNA.2002.1035072","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035072","url":null,"abstract":"In this paper we consider coupled quantum-dot cells, which are usually used for quantum-dot cellular automata (QCA), as a build unit to construct an analog cellular neural network. It is also shown how simple connection of few quantum-dot cells (even two of them) can cause the onset of chaotic oscillation only with small differences of polarizations and template between cells. An example of polarization evolution in two adjacent cells is reported together with proof of their chaotic behavior.","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":"115244494","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.2002.1035058
D. Bálya, C. Rekeczky, T. Roska
The unique possibility for reconstructing the first stage of the visual system on a programmable silicon chip is described. The developed mammalian retinal model can be implemented as an analogic algorithm running on a prototype complex cell cellular neural network processor. It enables the neuro-biological and vision research communities to study the wisdom of biological visual transformations design in real-time. The operating prototype complex-cell CNN-UM processor opens a new world for the engineering as well as the computational neuroscience communities. This paper demonstrates the first steps in this direction. Here we present the decomposition and scaling of one retinal channel as a hardware-level CNN-UM algorithm. The analogic algorithm consists of a series of different complex-cell CNN spatial-temporal dynamics, feasible on the recently fabricated prototype complex cell CNN-UM chip.
{"title":"Basic mammalian retinal effects on the prototype complex cell CNN universal machine","authors":"D. Bálya, C. Rekeczky, T. Roska","doi":"10.1109/CNNA.2002.1035058","DOIUrl":"https://doi.org/10.1109/CNNA.2002.1035058","url":null,"abstract":"The unique possibility for reconstructing the first stage of the visual system on a programmable silicon chip is described. The developed mammalian retinal model can be implemented as an analogic algorithm running on a prototype complex cell cellular neural network processor. It enables the neuro-biological and vision research communities to study the wisdom of biological visual transformations design in real-time. The operating prototype complex-cell CNN-UM processor opens a new world for the engineering as well as the computational neuroscience communities. This paper demonstrates the first steps in this direction. Here we present the decomposition and scaling of one retinal channel as a hardware-level CNN-UM algorithm. The analogic algorithm consists of a series of different complex-cell CNN spatial-temporal dynamics, feasible on the recently fabricated prototype complex cell CNN-UM chip.","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":"116004225","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}