Pub Date : 1994-12-18DOI: 10.1109/CNNA.1994.381703
D. Lim, G. Moschytz
An experimental monolithic implementation of a programmable cellular neural network (CNN) is reported. It overcomes some of the characteristics and restrictions inherent in CMOS VLSI technologies, and allows an arbitrarily large continuous-time analog CNN to be built up by modularly connecting CNN chips with a modest number of cells. The template values are step-wise programmable, with values chosen for functionality rather than according to conventional binary weighting. All external input, output and control signals are electrical and digital, so the CNN can be directly connected to a controller The design was carried out in a 1-micron n-well CMOS technology. Each cell occupies 0.4 mm/sup 2/, including all support circuitry; only one cell per chip was integrated in order to facilitate circuit testing. Measured CNN transients from a prototype 4/spl times/4 CNN, formed by connecting 16 one-cell chips are shown. The principal intended applications are the processing of acoustical signals and algorithm development.<>
{"title":"A programmable, modular CNN cell","authors":"D. Lim, G. Moschytz","doi":"10.1109/CNNA.1994.381703","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381703","url":null,"abstract":"An experimental monolithic implementation of a programmable cellular neural network (CNN) is reported. It overcomes some of the characteristics and restrictions inherent in CMOS VLSI technologies, and allows an arbitrarily large continuous-time analog CNN to be built up by modularly connecting CNN chips with a modest number of cells. The template values are step-wise programmable, with values chosen for functionality rather than according to conventional binary weighting. All external input, output and control signals are electrical and digital, so the CNN can be directly connected to a controller The design was carried out in a 1-micron n-well CMOS technology. Each cell occupies 0.4 mm/sup 2/, including all support circuitry; only one cell per chip was integrated in order to facilitate circuit testing. Measured CNN transients from a prototype 4/spl times/4 CNN, formed by connecting 16 one-cell chips are shown. The principal intended applications are the processing of acoustical signals and algorithm development.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115753942","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 : 1994-12-18DOI: 10.1109/CNNA.1994.381675
M. Balsi, N. Racina
Automatic recognition of tail signs placed on train tail cars is realized by CNN (cellular neural network) processing. This operation is required by Italian safety regulations and is currently done by a human operator. The CNN-based system proposed may already be sufficiently reliable and cheap to automate such a task.<>
{"title":"Automatic recognition of train tail signs using CNNs [Cellular neural networks]","authors":"M. Balsi, N. Racina","doi":"10.1109/CNNA.1994.381675","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381675","url":null,"abstract":"Automatic recognition of tail signs placed on train tail cars is realized by CNN (cellular neural network) processing. This operation is required by Italian safety regulations and is currently done by a human operator. The CNN-based system proposed may already be sufficiently reliable and cheap to automate such a task.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"13 7","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120821025","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 : 1994-12-18DOI: 10.1109/CNNA.1994.381672
A. Kellner, H. Magnussen, J. Nossek
Global Learning Algorithms presented in a companion paper are applied to practical classification and segmentation problems: Texture Classification and Texture Segmentation of artificial and natural textures, and Text Segmentation as a sub-problem of Page Layout Analysis. In all cases, DTCNN systems can solve the problem very well in spite of its only local interconnection structure.<>
{"title":"Texture classification, texture segmentation and text segmentation with discrete-time cellular neural networks","authors":"A. Kellner, H. Magnussen, J. Nossek","doi":"10.1109/CNNA.1994.381672","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381672","url":null,"abstract":"Global Learning Algorithms presented in a companion paper are applied to practical classification and segmentation problems: Texture Classification and Texture Segmentation of artificial and natural textures, and Text Segmentation as a sub-problem of Page Layout Analysis. In all cases, DTCNN systems can solve the problem very well in spite of its only local interconnection structure.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126104830","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 : 1994-12-18DOI: 10.1109/CNNA.1994.381661
Hanzhou Chen, Mingde Dai, Xinuan Wu
This paper studies bifurcation and chaos in discrete-time cellular neural networks (DTCNN), whose cells, similar to that in continuous-time CNN's, are locally coupled and whose output equations are logistic equations. The chaotic behavior of two types of DTCNN arrays, bounded and unbounded, is discussed respectively. While there is similarity between chaos of DTCNN's and that of globally coupled systems (Kaneko, 1990) DTCNN's differ from the latter in their bifurcation and statistical features due to their special locally coupled structure. Initial study on bifurcation and chaos in two-dimensional DTCNN arrays are presented in this paper with some interesting theoretical and practical problems proposed for our future research on this subject.<>
{"title":"Bifurcation and chaos in discrete-time cellular neural networks","authors":"Hanzhou Chen, Mingde Dai, Xinuan Wu","doi":"10.1109/CNNA.1994.381661","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381661","url":null,"abstract":"This paper studies bifurcation and chaos in discrete-time cellular neural networks (DTCNN), whose cells, similar to that in continuous-time CNN's, are locally coupled and whose output equations are logistic equations. The chaotic behavior of two types of DTCNN arrays, bounded and unbounded, is discussed respectively. While there is similarity between chaos of DTCNN's and that of globally coupled systems (Kaneko, 1990) DTCNN's differ from the latter in their bifurcation and statistical features due to their special locally coupled structure. Initial study on bifurcation and chaos in two-dimensional DTCNN arrays are presented in this paper with some interesting theoretical and practical problems proposed for our future research on this subject.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"428 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123274039","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 : 1994-12-18DOI: 10.1109/CNNA.1994.381687
Ta-lun Yang, Lin-Bao Yang, Guangyi Yang
In this paper, an algebraic criterion of the unconditional stability of the delayed cellular neural networks (DCNN) is presented. The criterion is necessary and sufficient, which gives the decision of the unconditional stability of the DCNN an elementary approach instead of the transcendental ones. Some examples are given to show how the criterion works in a simple way.<>
{"title":"On unconditional stability of the general delayed cellular neural networks","authors":"Ta-lun Yang, Lin-Bao Yang, Guangyi Yang","doi":"10.1109/CNNA.1994.381687","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381687","url":null,"abstract":"In this paper, an algebraic criterion of the unconditional stability of the delayed cellular neural networks (DCNN) is presented. The criterion is necessary and sufficient, which gives the decision of the unconditional stability of the DCNN an elementary approach instead of the transcendental ones. Some examples are given to show how the criterion works in a simple way.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125267283","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 : 1994-12-18DOI: 10.1109/CNNA.1994.381631
P. Arena, L. Fortuna, G. Manganaro, S. Spina
A new image processing technique based on cellular neural networks for improving the automatic classification of fruits (in particular, oranges) is introduced. It allows the digitised orange images to be processed in order to highlight some peculiarities of the fruits. In this way the following classification step is greatly simplified and improved. Moreover, the real-time processing characteristic of CNNs is a very advantageous point over the traditional computing resources commonly used in this kind of processing. The proposed task is accomplished by the choice of suitable templates in a simple CNN model. These templates are described and some examples are reported.<>
{"title":"CNN image processing for the automatic classification of oranges","authors":"P. Arena, L. Fortuna, G. Manganaro, S. Spina","doi":"10.1109/CNNA.1994.381631","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381631","url":null,"abstract":"A new image processing technique based on cellular neural networks for improving the automatic classification of fruits (in particular, oranges) is introduced. It allows the digitised orange images to be processed in order to highlight some peculiarities of the fruits. In this way the following classification step is greatly simplified and improved. Moreover, the real-time processing characteristic of CNNs is a very advantageous point over the traditional computing resources commonly used in this kind of processing. The proposed task is accomplished by the choice of suitable templates in a simple CNN model. These templates are described and some examples are reported.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114947268","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 : 1994-12-18DOI: 10.1109/CNNA.1994.381694
J. Nossek
The template coefficients (weights) of a CNN, which will give a desired performance, can either be found by design or by learning; "By design" means, that the desired function to be performed could be translated into a set of local dynamic rules, while "by learning" is based exclusively on pairs of input and corresponding output signals, the relationship of which may be by far too complicated for the explicit formulation of local rules. An overview of design and learning methods applicable to CNNs, which sometimes are not clearly distinguishable,is given here. Both technological constraints imposed by specific hardware implementation and practical constraints caused by the specific application and system embedding are influencing design and learning.<>
{"title":"Design and learning with cellular neural networks","authors":"J. Nossek","doi":"10.1109/CNNA.1994.381694","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381694","url":null,"abstract":"The template coefficients (weights) of a CNN, which will give a desired performance, can either be found by design or by learning; \"By design\" means, that the desired function to be performed could be translated into a set of local dynamic rules, while \"by learning\" is based exclusively on pairs of input and corresponding output signals, the relationship of which may be by far too complicated for the explicit formulation of local rules. An overview of design and learning methods applicable to CNNs, which sometimes are not clearly distinguishable,is given here. Both technological constraints imposed by specific hardware implementation and practical constraints caused by the specific application and system embedding are influencing design and learning.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"219 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116384427","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 : 1994-12-18DOI: 10.1109/CNNA.1994.381699
Xiang-Zhu Huang, Ta-lun Yang, Lin-Bao Yang
In this paper, the stability of the time-variant delayed cellular neural networks (TVDCNN) is presented. The effects of the time-variant templates, the inner parameters of cells and other parameters of changing circumstances to the stability of the DCNNs are studied. Some sufficient conditions concerning the bound of delay are presented to ensure the stability of the TVDCNNs.<>
{"title":"On stability of the time-variant delayed cellular neural networks","authors":"Xiang-Zhu Huang, Ta-lun Yang, Lin-Bao Yang","doi":"10.1109/CNNA.1994.381699","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381699","url":null,"abstract":"In this paper, the stability of the time-variant delayed cellular neural networks (TVDCNN) is presented. The effects of the time-variant templates, the inner parameters of cells and other parameters of changing circumstances to the stability of the DCNNs are studied. Some sufficient conditions concerning the bound of delay are presented to ensure the stability of the TVDCNNs.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122133284","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 : 1994-12-18DOI: 10.1109/CNNA.1994.381700
E. Pessa, M.P. Penna
In this contribution neuronic cellular automata, a particular subcase of DTCNN, are studied from the point of view of the influence of the distribution of their connection weights on their dynamic behavior and on their spatial correlation length.<>
{"title":"Local and global connectivity in neuronic cellular automata","authors":"E. Pessa, M.P. Penna","doi":"10.1109/CNNA.1994.381700","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381700","url":null,"abstract":"In this contribution neuronic cellular automata, a particular subcase of DTCNN, are studied from the point of view of the influence of the distribution of their connection weights on their dynamic behavior and on their spatial correlation length.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"235 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132274087","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 : 1994-12-18DOI: 10.1109/CNNA.1994.381647
M. Grimaila, J. P. de Gyvez
A CAD tool based on SPICE macromodels to simulate simplified faulty, circuit realizations of a fully programmable, two dimensional cellular neural network (CNN) is presented. The models can be easily adapted to match the electrical parameters of real circuit implementations. Generic macromodels for both current mode and voltage mode CNNs are provided. The macromodels not only simulate the conceptual CNN cell, but also provide the capability to model actual CNN architectures and their nonidealities. Moreover, macromodeling provides the capability to determine the effect of parameter variation on the operation of the CNN efficiently without the need for computationally expensive, exhaustive circuit simulations. We have used the CNN macromodels to develop robust testing strategies for detecting faults in VLSI implementations of CNN arrays. Three fault cases are introduced into a CNN array to provide insight to the usefulness of macromodeling.<>
{"title":"A macromodel fault generator for cellular neural networks","authors":"M. Grimaila, J. P. de Gyvez","doi":"10.1109/CNNA.1994.381647","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381647","url":null,"abstract":"A CAD tool based on SPICE macromodels to simulate simplified faulty, circuit realizations of a fully programmable, two dimensional cellular neural network (CNN) is presented. The models can be easily adapted to match the electrical parameters of real circuit implementations. Generic macromodels for both current mode and voltage mode CNNs are provided. The macromodels not only simulate the conceptual CNN cell, but also provide the capability to model actual CNN architectures and their nonidealities. Moreover, macromodeling provides the capability to determine the effect of parameter variation on the operation of the CNN efficiently without the need for computationally expensive, exhaustive circuit simulations. We have used the CNN macromodels to develop robust testing strategies for detecting faults in VLSI implementations of CNN arrays. Three fault cases are introduced into a CNN array to provide insight to the usefulness of macromodeling.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"92 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114899778","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：