Pub Date : 1991-08-15DOI: 10.1109/ICNN.1991.163365
P. K. Simpson
A feedforward neural network classifier that uses min-max vector pairs to define classes is described. This two-layer neural network utilizes a supervised learning rule to build a set of classes. Each node in the output layer of the network represents a class. During recall each class node produces an output value that represents the degree to which the input pattern fits within the represented classes. This fuzzy neural network is ideally suited to applications that have very little data available to define classes. The author provides a brief overview of fuzzy sets and fuzzy pattern classification, a description of fuzzy min-max classification and its neural network implementation, and an example of the classification operation.<>
{"title":"Fuzzy min-max classification with neural networks","authors":"P. K. Simpson","doi":"10.1109/ICNN.1991.163365","DOIUrl":"https://doi.org/10.1109/ICNN.1991.163365","url":null,"abstract":"A feedforward neural network classifier that uses min-max vector pairs to define classes is described. This two-layer neural network utilizes a supervised learning rule to build a set of classes. Each node in the output layer of the network represents a class. During recall each class node produces an output value that represents the degree to which the input pattern fits within the represented classes. This fuzzy neural network is ideally suited to applications that have very little data available to define classes. The author provides a brief overview of fuzzy sets and fuzzy pattern classification, a description of fuzzy min-max classification and its neural network implementation, and an example of the classification operation.<<ETX>>","PeriodicalId":296300,"journal":{"name":"[1991 Proceedings] IEEE Conference on Neural Networks for Ocean Engineering","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134383404","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 : 1991-08-15DOI: 10.1109/ICNN.1991.163347
M. Dzwonczyk, M. Busa, J. T. Sims, T. Daud
An integrated neurocomputing architecture developed for deployable, real-time pattern recognition applications is described. This architecture, called INCA, consists of a fully parallel, analog electronic, feedforward neural network coupled with a conventional microprocessor system. The first generation system, INCA/1, is currently under construction and employs existing analog neural network building block chips, with an off-the-shelf single-board computer. The proof-of-concept application for INCA/1 is the automatic detection of targets in sidescan sonar images. Preliminary simulations of the network, which account for some of the characteristics of the physical electronics, have shown excellent performance on real data without preprocessing.<>
{"title":"An integrated neurocomputing architecture for side-scan sonar target detection","authors":"M. Dzwonczyk, M. Busa, J. T. Sims, T. Daud","doi":"10.1109/ICNN.1991.163347","DOIUrl":"https://doi.org/10.1109/ICNN.1991.163347","url":null,"abstract":"An integrated neurocomputing architecture developed for deployable, real-time pattern recognition applications is described. This architecture, called INCA, consists of a fully parallel, analog electronic, feedforward neural network coupled with a conventional microprocessor system. The first generation system, INCA/1, is currently under construction and employs existing analog neural network building block chips, with an off-the-shelf single-board computer. The proof-of-concept application for INCA/1 is the automatic detection of targets in sidescan sonar images. Preliminary simulations of the network, which account for some of the characteristics of the physical electronics, have shown excellent performance on real data without preprocessing.<<ETX>>","PeriodicalId":296300,"journal":{"name":"[1991 Proceedings] IEEE Conference on Neural Networks for Ocean Engineering","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132950886","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 : 1991-08-15DOI: 10.1109/ICNN.1991.163329
A. Russo
The author describes a neural network system that recognizes seven different types of passive sonar signals from their characteristic shapes. The system has a preprocessor for signal detection and symbolic representation, a bank of three highly constrained feedforward neural networks for recognition, and a postprocessor for network interpretation and performance adjustment. The preprocessor uses image processing and morphological techniques to extract and track energy, and converts each detected signal into a chain code. The chain code is passed to an ensemble of three independent neural networks, each of which votes on the signal's type. The system's performance on 1400 unseen test signals was an adjustable 93% overall correct recognition rate, 5% error rate, and 2% rejection rate.<>
{"title":"Constrained neural networks for recognition of passive sonar signals using shape","authors":"A. Russo","doi":"10.1109/ICNN.1991.163329","DOIUrl":"https://doi.org/10.1109/ICNN.1991.163329","url":null,"abstract":"The author describes a neural network system that recognizes seven different types of passive sonar signals from their characteristic shapes. The system has a preprocessor for signal detection and symbolic representation, a bank of three highly constrained feedforward neural networks for recognition, and a postprocessor for network interpretation and performance adjustment. The preprocessor uses image processing and morphological techniques to extract and track energy, and converts each detected signal into a chain code. The chain code is passed to an ensemble of three independent neural networks, each of which votes on the signal's type. The system's performance on 1400 unseen test signals was an adjustable 93% overall correct recognition rate, 5% error rate, and 2% rejection rate.<<ETX>>","PeriodicalId":296300,"journal":{"name":"[1991 Proceedings] IEEE Conference on Neural Networks for Ocean Engineering","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125598254","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 : 1991-08-15DOI: 10.1109/ICNN.1991.163348
J. Stover, R. E. Gibson
The authors describe a structure for establishing the existence of fuzzy properties or patterns from a hierarchical sequence of fuzzy subproperties. The structure, referred to as a continuous inference network, consists of nodes that combine information of differing degrees of significance as well as differing degrees of existence. A discussion of characteristics of node transfer functions needed in the autonomous underwater vehicle/remotely operated vehicle (AUV/ROV) controller software and integration with neural network subsystems is included.<>
{"title":"Continuous inference networks for autonomous systems","authors":"J. Stover, R. E. Gibson","doi":"10.1109/ICNN.1991.163348","DOIUrl":"https://doi.org/10.1109/ICNN.1991.163348","url":null,"abstract":"The authors describe a structure for establishing the existence of fuzzy properties or patterns from a hierarchical sequence of fuzzy subproperties. The structure, referred to as a continuous inference network, consists of nodes that combine information of differing degrees of significance as well as differing degrees of existence. A discussion of characteristics of node transfer functions needed in the autonomous underwater vehicle/remotely operated vehicle (AUV/ROV) controller software and integration with neural network subsystems is included.<<ETX>>","PeriodicalId":296300,"journal":{"name":"[1991 Proceedings] IEEE Conference on Neural Networks for Ocean Engineering","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124871779","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 : 1991-08-15DOI: 10.1109/ICNN.1991.163325
Joydeep Ghosh, S. Chakravarthy, Y. Shin, C. Chu, L. Deuser, S. Beck, R. Still, J. Whiteley
Two kernel networks are presented for the classification of short-duration acoustic signals characterized by wavelet coefficients and signal duration. These networks combine the positive features of exemplar-based classifiers such as the learned vector quantization method and kernel classifiers using radial basis functions. Results on the DARPA Data Set 1 show that these networks compare favorably with other classification techniques, with almost 100% accuracy achievable in identifying test signals that are similar to the training signals. A method of combining the outputs of several classifiers to yield a more accurate labeling is proposed based on the interpretation of network outputs as approximating posterior class probabilities. The authors also provide a technique for recognizing deviant signals and false alarms.<>
{"title":"Adaptive kernel classifiers for short-duration oceanic signals","authors":"Joydeep Ghosh, S. Chakravarthy, Y. Shin, C. Chu, L. Deuser, S. Beck, R. Still, J. Whiteley","doi":"10.1109/ICNN.1991.163325","DOIUrl":"https://doi.org/10.1109/ICNN.1991.163325","url":null,"abstract":"Two kernel networks are presented for the classification of short-duration acoustic signals characterized by wavelet coefficients and signal duration. These networks combine the positive features of exemplar-based classifiers such as the learned vector quantization method and kernel classifiers using radial basis functions. Results on the DARPA Data Set 1 show that these networks compare favorably with other classification techniques, with almost 100% accuracy achievable in identifying test signals that are similar to the training signals. A method of combining the outputs of several classifiers to yield a more accurate labeling is proposed based on the interpretation of network outputs as approximating posterior class probabilities. The authors also provide a technique for recognizing deviant signals and false alarms.<<ETX>>","PeriodicalId":296300,"journal":{"name":"[1991 Proceedings] IEEE Conference on Neural Networks for Ocean Engineering","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124900564","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 : 1991-08-15DOI: 10.1109/ICNN.1991.163328
D.J. Shazeer, M. Bello
The authors describe the use of multilayer perceptrons to solve the problem of distinguishing mine-like objects from clutter. Three increasingly sophisticated and effective approaches were applied against difficult side scan sonar imagery containing a highly cluttered and variable environment. Performances of the three approaches are compared using receiver operating curves (ROCs). Comparisons show that one can achieve a detection rate of 0.97 for a 0.01 false alarm rate. A subset of the networks have been demonstrated on special purpose hardware to run in real time.<>
{"title":"Minehunting with multi-layer perceptrons","authors":"D.J. Shazeer, M. Bello","doi":"10.1109/ICNN.1991.163328","DOIUrl":"https://doi.org/10.1109/ICNN.1991.163328","url":null,"abstract":"The authors describe the use of multilayer perceptrons to solve the problem of distinguishing mine-like objects from clutter. Three increasingly sophisticated and effective approaches were applied against difficult side scan sonar imagery containing a highly cluttered and variable environment. Performances of the three approaches are compared using receiver operating curves (ROCs). Comparisons show that one can achieve a detection rate of 0.97 for a 0.01 false alarm rate. A subset of the networks have been demonstrated on special purpose hardware to run in real time.<<ETX>>","PeriodicalId":296300,"journal":{"name":"[1991 Proceedings] IEEE Conference on Neural Networks for Ocean Engineering","volume":"552 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123265543","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 : 1991-08-15DOI: 10.1109/ICNN.1991.163369
S. Sin, R. de Figueiredo
An evolutionary design methodology for neural networks based on the theory of optimal interpolation, (OI) is presented. A limited application of the OI net to the problems of localization and classification of acoustic transients is discussed. The modified recursive least squares (RLS) learning algorithm presented provides an avenue for the acquisition of an appropriate neural network configuration to solve a given pattern classification problem. The authors show that both OI and the back-propagation (BP) of comparable configurations perform satisfactorily in the simulations. The RLS OI method is preferred, however, because BP would occasionally run into some local minima and convergence could be very slow for the more complex decision boundaries between classes. The authors demonstrate that the OI net is particularly suited for application to the localization and classification of acoustic transients.<>
{"title":"A new design methodology for optimal interpolative neural networks with application to the localization and classification of acoustic transients","authors":"S. Sin, R. de Figueiredo","doi":"10.1109/ICNN.1991.163369","DOIUrl":"https://doi.org/10.1109/ICNN.1991.163369","url":null,"abstract":"An evolutionary design methodology for neural networks based on the theory of optimal interpolation, (OI) is presented. A limited application of the OI net to the problems of localization and classification of acoustic transients is discussed. The modified recursive least squares (RLS) learning algorithm presented provides an avenue for the acquisition of an appropriate neural network configuration to solve a given pattern classification problem. The authors show that both OI and the back-propagation (BP) of comparable configurations perform satisfactorily in the simulations. The RLS OI method is preferred, however, because BP would occasionally run into some local minima and convergence could be very slow for the more complex decision boundaries between classes. The authors demonstrate that the OI net is particularly suited for application to the localization and classification of acoustic transients.<<ETX>>","PeriodicalId":296300,"journal":{"name":"[1991 Proceedings] IEEE Conference on Neural Networks for Ocean Engineering","volume":"PP 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126444697","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 : 1991-08-15DOI: 10.1109/ICNN.1991.163358
D. Montana, K. Theriault
The authors developed systems for detection and classification of acoustic transients. Here, the authors describe the insights and interim results so far obtained. The general processing architecture used is presented. They examine the major difficulties of this problem as compared with simpler pattern classification problems. They discuss a set of experiments which support many of the development and design guidelines. They describe what these guidelines are and provide further justification for their importance.<>
{"title":"Neural-network-based classification of acoustic transients","authors":"D. Montana, K. Theriault","doi":"10.1109/ICNN.1991.163358","DOIUrl":"https://doi.org/10.1109/ICNN.1991.163358","url":null,"abstract":"The authors developed systems for detection and classification of acoustic transients. Here, the authors describe the insights and interim results so far obtained. The general processing architecture used is presented. They examine the major difficulties of this problem as compared with simpler pattern classification problems. They discuss a set of experiments which support many of the development and design guidelines. They describe what these guidelines are and provide further justification for their importance.<<ETX>>","PeriodicalId":296300,"journal":{"name":"[1991 Proceedings] IEEE Conference on Neural Networks for Ocean Engineering","volume":"153 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127115484","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 : 1991-08-15DOI: 10.1109/ICNN.1991.163330
S. Speidel
A computing architecture is being produced that automates primitive and schemea-based streaming of sounds and thereby achieves better real-time, in-situ analyses of complicated sonar scenes. The computational models are called the neural beamformers (NBFs). A brief qualitative overview of three beamformers is given: the crossbar beamformer is based on the Hopfield crossbar circuit; the multivector beamformer is related to Kohonen feature map learning; and the neurobionic beamformer is really a network of beamformers and combines elements of the other two beamformers. In experiments using an array of microphones operated in a laboratory room, an NBF was able to locate a sound source while exhibiting tolerance to sounds arriving at the array via a reflected path once the processing had seen the onset of the direct path excitation from the source.<>
{"title":"Sonar scene analysis using neurobionic sound segregation","authors":"S. Speidel","doi":"10.1109/ICNN.1991.163330","DOIUrl":"https://doi.org/10.1109/ICNN.1991.163330","url":null,"abstract":"A computing architecture is being produced that automates primitive and schemea-based streaming of sounds and thereby achieves better real-time, in-situ analyses of complicated sonar scenes. The computational models are called the neural beamformers (NBFs). A brief qualitative overview of three beamformers is given: the crossbar beamformer is based on the Hopfield crossbar circuit; the multivector beamformer is related to Kohonen feature map learning; and the neurobionic beamformer is really a network of beamformers and combines elements of the other two beamformers. In experiments using an array of microphones operated in a laboratory room, an NBF was able to locate a sound source while exhibiting tolerance to sounds arriving at the array via a reflected path once the processing had seen the onset of the direct path excitation from the source.<<ETX>>","PeriodicalId":296300,"journal":{"name":"[1991 Proceedings] IEEE Conference on Neural Networks for Ocean Engineering","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126781849","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 : 1991-08-15DOI: 10.1109/ICNN.1991.163349
N. Seube
The authors present two original learning rules for control and compare their performance in the control of an autonomous underwater vehicle. The problem of tracking a reference trajectory with neural controllers is also investigated. The authors discuss the adaptive features of neural networks for control. It is experimentally and theoretically shown that one of the learning rules proposed can perform accurate tracking control in a nonlinear system theory, which explains regulation mechanisms of state-constrained control systems. Numerical results are presented for the tracking control of the dolphin 3 K vehicle.<>
{"title":"Neural network learning rules for control: application to AUV tracking control","authors":"N. Seube","doi":"10.1109/ICNN.1991.163349","DOIUrl":"https://doi.org/10.1109/ICNN.1991.163349","url":null,"abstract":"The authors present two original learning rules for control and compare their performance in the control of an autonomous underwater vehicle. The problem of tracking a reference trajectory with neural controllers is also investigated. The authors discuss the adaptive features of neural networks for control. It is experimentally and theoretically shown that one of the learning rules proposed can perform accurate tracking control in a nonlinear system theory, which explains regulation mechanisms of state-constrained control systems. Numerical results are presented for the tracking control of the dolphin 3 K vehicle.<<ETX>>","PeriodicalId":296300,"journal":{"name":"[1991 Proceedings] IEEE Conference on Neural Networks for Ocean Engineering","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126687432","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：