Pub Date : 1994-12-18DOI: 10.1109/CNNA.1994.381655
Z. Galias, J. Nossek
Summary form only given. We study the possibilities of suppressing chaotic behaviour of the three-cell cellular neural network. We present the laboratory environment and experimental results of stabilization of one of the existing unstable periodic orbits, by means of applying small periodic perturbations to one of the circuit parameters. The results obtained are promising. The data acquisition and identification part work correctly. Based on time series obtained from the real process, we have found several unstable periodic orbits and their parameters necessary for the control. We have performed a number of control experiments. We have measured the performance of the system and noticed that the trajectory remains longer in the neighbourhood of the stabilized periodic orbit in the case when the control is active. We conclude that the control method is sensitive to noise and accuracy of the computed parameters of the stabilized periodic orbit. We believe that with some modifications a successful control is possible.<>
{"title":"Control of a real chaotic cellular neural network","authors":"Z. Galias, J. Nossek","doi":"10.1109/CNNA.1994.381655","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381655","url":null,"abstract":"Summary form only given. We study the possibilities of suppressing chaotic behaviour of the three-cell cellular neural network. We present the laboratory environment and experimental results of stabilization of one of the existing unstable periodic orbits, by means of applying small periodic perturbations to one of the circuit parameters. The results obtained are promising. The data acquisition and identification part work correctly. Based on time series obtained from the real process, we have found several unstable periodic orbits and their parameters necessary for the control. We have performed a number of control experiments. We have measured the performance of the system and noticed that the trajectory remains longer in the neighbourhood of the stabilized periodic orbit in the case when the control is active. We conclude that the control method is sensitive to noise and accuracy of the computed parameters of the stabilized periodic orbit. We believe that with some modifications a successful control is possible.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"268 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":"131627777","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.1994.381658
P. Civalleri, M. Gilli
The propagation phenomena occurring in cellular neural networks (CNN's) described by a one-dimensional template are investigated by using a spectral technique. The CNN is represented as a scalar Lur'e system to which a suitable extension of the describing function technique is applied. It is shown that the method yields results that are in very good agreement with those observed by the time-simulation of the system.<>
{"title":"Propagation phenomena in cellular neural networks","authors":"P. Civalleri, M. Gilli","doi":"10.1109/CNNA.1994.381658","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381658","url":null,"abstract":"The propagation phenomena occurring in cellular neural networks (CNN's) described by a one-dimensional template are investigated by using a spectral technique. The CNN is represented as a scalar Lur'e system to which a suitable extension of the describing function technique is applied. It is shown that the method yields results that are in very good agreement with those observed by the time-simulation of the system.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"120 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":"116523003","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.1994.381630
S. Schwarz
The paper discusses the detection of weaknesses and defects on photolithographic masks by cellular neural networks. The detection by cellular neural networks is performed with the advantages of their massive parallel architecture. First a survey is given of actual methods for the detections of weaknesses and defects. Then the relations between the structures of the mask layouts and the real structures of the masks are defined by local design rules. These local design rules can also indirectly be used to detect most weaknesses and defects. After that, the design of the operators for the detection of weaknesses and defects are executed on the basis of the local design rules, using the method of Galias that is practicable by cellular neural networks. Then some examples of weakness and defect detections on real mask images by cellular neural networks are presented. Finally the results and future aims are discussed.<>
{"title":"Detection of defects on photolithographic masks by cellular neural networks","authors":"S. Schwarz","doi":"10.1109/CNNA.1994.381630","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381630","url":null,"abstract":"The paper discusses the detection of weaknesses and defects on photolithographic masks by cellular neural networks. The detection by cellular neural networks is performed with the advantages of their massive parallel architecture. First a survey is given of actual methods for the detections of weaknesses and defects. Then the relations between the structures of the mask layouts and the real structures of the masks are defined by local design rules. These local design rules can also indirectly be used to detect most weaknesses and defects. After that, the design of the operators for the detection of weaknesses and defects are executed on the basis of the local design rules, using the method of Galias that is practicable by cellular neural networks. Then some examples of weakness and defect detections on real mask images by cellular neural networks are presented. Finally the results and future aims are discussed.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"105 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":"124787643","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.1994.381698
J.J. Szczyrek, S. Jankowski
A new class of cloning templates in nonreciprocal cellular neural networks is proposed. Basing on the opposite sign template system introduced by (Chua and Roska, 1990) the stability of the wider class of CNN which are nonsymmetrical is considered. No assumptions about size of neighborhood and topological structure of CNN (e.g. dimension) follow to generalize the results.<>
提出了一类新的非互易细胞神经网络克隆模板。在(Chua and Roska, 1990)引入的对号模板系统的基础上，考虑了非对称广义CNN的稳定性。没有对CNN的邻域大小和拓扑结构(例如维数)的假设来推广结果。
{"title":"A class of asymmetrical templates in cellular neural networks","authors":"J.J. Szczyrek, S. Jankowski","doi":"10.1109/CNNA.1994.381698","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381698","url":null,"abstract":"A new class of cloning templates in nonreciprocal cellular neural networks is proposed. Basing on the opposite sign template system introduced by (Chua and Roska, 1990) the stability of the wider class of CNN which are nonsymmetrical is considered. No assumptions about size of neighborhood and topological structure of CNN (e.g. dimension) follow to generalize the results.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"74 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":"124995946","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.1994.381640
S. Espejo, Á. Rodríguez-Vázquez, R. Domínguez-Castro, R. Carmona
Stability and convergency results are reported for a modified continuous-time CNN model. The signal range of the state variables is equal to the unitary interval, independently of the application, Stability and convergency properties are similar to those of the original model and, for given templates and offset coefficients. The results are generally identical. In addition, robustness and area-efficiency of VLSI implementations are significantly advantageous.<>
{"title":"Convergence and stability of the FSR CNN model","authors":"S. Espejo, Á. Rodríguez-Vázquez, R. Domínguez-Castro, R. Carmona","doi":"10.1109/CNNA.1994.381640","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381640","url":null,"abstract":"Stability and convergency results are reported for a modified continuous-time CNN model. The signal range of the state variables is equal to the unitary interval, independently of the application, Stability and convergency properties are similar to those of the original model and, for given templates and offset coefficients. The results are generally identical. In addition, robustness and area-efficiency of VLSI implementations are significantly advantageous.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"24 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":"114169346","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.1994.381688
C. Guzelis, S. Karamahmut
A supervised learning algorithm for obtaining the template coefficients in completely stable cellular neural networks (CNNs) is presented. The proposed algorithm resembles the well-known perceptron learning algorithm and hence is called as recurrent perceptron learning algorithm (RPLA) as applied to a dynamical network, CNN. The RPLA can be described as the following set of rules: (i) increase each feedback template coefficient which defines the connection to a mismatching cell from its neighbor whose steady-state output is same with the mismatching cell's desired output. On the contrary, decrease each feedback template coefficient which defines the connection to a mismatching cell from its neighbor whose steady-state is different from the mismatching cell's desired output. (ii) Change the input template coefficients according to the rule stated in (i) by only replacing the word of "neighbor" with "input". (iii) Retain the template coefficients unchanged if the actual outputs match the desired outputs. The proposed algorithm RPLA has been applied for training CNNs to perform several image processing tasks such as edge detection, hole filling and corner detection. The performance of the templates obtained for the chosen input-(desired)output training pairs has been tested on a set of images which are different from the input images used in the training phase.<>
{"title":"Recurrent perceptron learning algorithm for completely stable cellular neural networks","authors":"C. Guzelis, S. Karamahmut","doi":"10.1109/CNNA.1994.381688","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381688","url":null,"abstract":"A supervised learning algorithm for obtaining the template coefficients in completely stable cellular neural networks (CNNs) is presented. The proposed algorithm resembles the well-known perceptron learning algorithm and hence is called as recurrent perceptron learning algorithm (RPLA) as applied to a dynamical network, CNN. The RPLA can be described as the following set of rules: (i) increase each feedback template coefficient which defines the connection to a mismatching cell from its neighbor whose steady-state output is same with the mismatching cell's desired output. On the contrary, decrease each feedback template coefficient which defines the connection to a mismatching cell from its neighbor whose steady-state is different from the mismatching cell's desired output. (ii) Change the input template coefficients according to the rule stated in (i) by only replacing the word of \"neighbor\" with \"input\". (iii) Retain the template coefficients unchanged if the actual outputs match the desired outputs. The proposed algorithm RPLA has been applied for training CNNs to perform several image processing tasks such as edge detection, hole filling and corner detection. The performance of the templates obtained for the chosen input-(desired)output training pairs has been tested on a set of images which are different from the input images used in the training phase.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"32 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":"132904853","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.1994.381645
P. Kinget, M. Steyaert
In this paper a method for the evaluation of the static robustness of cellular neural network (CNN) templates is proposed. From this evaluation the circuit accuracy specifications for a VLSI implementation can be derived which allows the designer to optimize the performance. Moreover, from this evaluation method guidelines for robust template designs can be derived and parameter testing templates can be developed.<>
{"title":"Evaluation of CNN template robustness towards VLSI implementation","authors":"P. Kinget, M. Steyaert","doi":"10.1109/CNNA.1994.381645","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381645","url":null,"abstract":"In this paper a method for the evaluation of the static robustness of cellular neural network (CNN) templates is proposed. From this evaluation the circuit accuracy specifications for a VLSI implementation can be derived which allows the designer to optimize the performance. Moreover, from this evaluation method guidelines for robust template designs can be derived and parameter testing templates can be developed.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"615 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":"133732327","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.1994.381689
H. Magnussen, G. Papoutsis, J. Nossek
The SGN-type nonlinearity of a standard discrete-time cellular neural network (DTCNN) is replaced by a smooth, sigmoidal nonlinearity with variable gain. Therefore, the resulting dynamical system is fully differentiable. Bounds on gain of the sigmoidal function are given, so that the new smooth system approximates the standard DTCNN within certain limits. A learning algorithm is proposed, which finds the template parameters for the standard DTCNN by gradually increasing the gain of the sigmoidal function.<>
{"title":"Continuation-based learning algorithm for discrete-time cellular neural networks","authors":"H. Magnussen, G. Papoutsis, J. Nossek","doi":"10.1109/CNNA.1994.381689","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381689","url":null,"abstract":"The SGN-type nonlinearity of a standard discrete-time cellular neural network (DTCNN) is replaced by a smooth, sigmoidal nonlinearity with variable gain. Therefore, the resulting dynamical system is fully differentiable. Bounds on gain of the sigmoidal function are given, so that the new smooth system approximates the standard DTCNN within certain limits. A learning algorithm is proposed, which finds the template parameters for the standard DTCNN by gradually increasing the gain of the sigmoidal function.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"13 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":"134164963","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.1994.381680
H. Mizutani
Multilayered neural networks may provide an effective way to make general associative neural networks but it is difficult to fabricate such a network. So multilayered CNNs are important. A new mapping method and a modified backpropagation method for the multilayered CNN is presented.<>
{"title":"A new learning method for multilayered cellular neural networks","authors":"H. Mizutani","doi":"10.1109/CNNA.1994.381680","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381680","url":null,"abstract":"Multilayered neural networks may provide an effective way to make general associative neural networks but it is difficult to fabricate such a network. So multilayered CNNs are important. A new mapping method and a modified backpropagation method for the multilayered CNN is presented.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"7 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":"133913931","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.1994.381667
Á. Zarándy, F. Werblin, T. Roska, L. Chua
Novel types of analogic algorithms, using spatio-temporal CNN (cellular nonlinear/neural networks) operations are introduced. These algorithms make complex decisions in images without reading out the CNN chip. This makes them extremely time, area, and power effective. Two crucial effects are emphasized: diffusion type templates are applied during a finite time interval and local logic operates within well defined parts (patches) in the image plane. Hence, a new type of pattern recognition algorithm is introduced. The technique is demonstrated on an example. In our example we are dealing with an actual problem: how to avoid the counterfeiting on color copiers.<>
{"title":"Novel types of analogic CNN algorithms for recognizing bank-notes","authors":"Á. Zarándy, F. Werblin, T. Roska, L. Chua","doi":"10.1109/CNNA.1994.381667","DOIUrl":"https://doi.org/10.1109/CNNA.1994.381667","url":null,"abstract":"Novel types of analogic algorithms, using spatio-temporal CNN (cellular nonlinear/neural networks) operations are introduced. These algorithms make complex decisions in images without reading out the CNN chip. This makes them extremely time, area, and power effective. Two crucial effects are emphasized: diffusion type templates are applied during a finite time interval and local logic operates within well defined parts (patches) in the image plane. Hence, a new type of pattern recognition algorithm is introduced. The technique is demonstrated on an example. In our example we are dealing with an actual problem: how to avoid the counterfeiting on color copiers.<<ETX>>","PeriodicalId":248898,"journal":{"name":"Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94)","volume":"27 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":"114717177","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：