Pub Date : 1996-12-09DOI: 10.1109/CYBVIS.1996.629449
P. R. Oliveira, R. Romero
Principal component analysis (PCA), also called Karhunen-Loeve transform, is a statistical method for multivariate data analysis that can be used in particular to reduce the data set being considered. There are two approaches for performing PCA. The first utilizes the classical statistical method and the other, artificial neural networks. In this paper, neural networks that performing PCA are presented and used to realize tomographic image compression. The results obtained are compared to that obtained by using JPEG compression standard technique and show the usefulness of neural networks for performing image compression.
{"title":"A comparision between PCA neural networks and the JPEG standard for performing image compression","authors":"P. R. Oliveira, R. Romero","doi":"10.1109/CYBVIS.1996.629449","DOIUrl":"https://doi.org/10.1109/CYBVIS.1996.629449","url":null,"abstract":"Principal component analysis (PCA), also called Karhunen-Loeve transform, is a statistical method for multivariate data analysis that can be used in particular to reduce the data set being considered. There are two approaches for performing PCA. The first utilizes the classical statistical method and the other, artificial neural networks. In this paper, neural networks that performing PCA are presented and used to realize tomographic image compression. The results obtained are compared to that obtained by using JPEG compression standard technique and show the usefulness of neural networks for performing image compression.","PeriodicalId":103287,"journal":{"name":"Proceedings II Workshop on Cybernetic Vision","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122617908","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 : 1996-12-09DOI: 10.1109/CYBVIS.1996.629471
C. Formigoni, C. A. Pela, J. Colafêmina
We investigated the interaction of the vestibular nucleus (VN) with the nucleus of optic tract (NOT) for gaze stabilization. Observer movement changes the image that comes into the retina and the NOT vestibular interaction can be responsible for gaze stabilization. To verify this behavior the subjects were stimulated with visual patterns at 85 deg/sec. We waited 10 minutes for the eye's accommodation because the room has a controlled light intensity separated by 11.5 lux. We observed that an angular displacement and a gain of a slow phase component of the nystagmus occurred when visual pathways are "on" and a decrease in amplitude of the signal showing that the NOT modulates the vestibular nucleus (VN).
{"title":"Interactions of the nucleus of optic tract and vestibular system with gaze stabilization","authors":"C. Formigoni, C. A. Pela, J. Colafêmina","doi":"10.1109/CYBVIS.1996.629471","DOIUrl":"https://doi.org/10.1109/CYBVIS.1996.629471","url":null,"abstract":"We investigated the interaction of the vestibular nucleus (VN) with the nucleus of optic tract (NOT) for gaze stabilization. Observer movement changes the image that comes into the retina and the NOT vestibular interaction can be responsible for gaze stabilization. To verify this behavior the subjects were stimulated with visual patterns at 85 deg/sec. We waited 10 minutes for the eye's accommodation because the room has a controlled light intensity separated by 11.5 lux. We observed that an angular displacement and a gain of a slow phase component of the nystagmus occurred when visual pathways are \"on\" and a decrease in amplitude of the signal showing that the NOT modulates the vestibular nucleus (VN).","PeriodicalId":103287,"journal":{"name":"Proceedings II Workshop on Cybernetic Vision","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123911911","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 : 1996-12-09DOI: 10.1109/CYBVIS.1996.629443
A. de Carvalho, D. Bisset, M. Fairhurst
This paper presents and analyses distinct learning strategies which have been used by GSN/sup f/ architectures. Sharing the common feature of being one-shot learning, these strategies achieve different performances as key parameters are changed. These algorithms are evaluated against each other by taking into account the training time, saturation rates, learning conflict rates and recognition performance.
{"title":"Training algorithms for GSN/sup f/ neural networks","authors":"A. de Carvalho, D. Bisset, M. Fairhurst","doi":"10.1109/CYBVIS.1996.629443","DOIUrl":"https://doi.org/10.1109/CYBVIS.1996.629443","url":null,"abstract":"This paper presents and analyses distinct learning strategies which have been used by GSN/sup f/ architectures. Sharing the common feature of being one-shot learning, these strategies achieve different performances as key parameters are changed. These algorithms are evaluated against each other by taking into account the training time, saturation rates, learning conflict rates and recognition performance.","PeriodicalId":103287,"journal":{"name":"Proceedings II Workshop on Cybernetic Vision","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127285009","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 : 1996-12-09DOI: 10.1109/CYBVIS.1996.629463
E. Neves, A. Gonzaga, A. Slaets
Neural networks present an alternative approach for the character recognition problem. This paper describes the development of a recognition system of multi-font character using topological feature extraction to recognize capital isolated letters. By properly specifying a set of features such as vertical, horizontal, and slant strokes, curvature, open and closed areas, called here "fundamental features", the recognition was performed using a backpropagation neural network.
{"title":"A multi-font character recognition based on its fundamental features by artificial neural networks","authors":"E. Neves, A. Gonzaga, A. Slaets","doi":"10.1109/CYBVIS.1996.629463","DOIUrl":"https://doi.org/10.1109/CYBVIS.1996.629463","url":null,"abstract":"Neural networks present an alternative approach for the character recognition problem. This paper describes the development of a recognition system of multi-font character using topological feature extraction to recognize capital isolated letters. By properly specifying a set of features such as vertical, horizontal, and slant strokes, curvature, open and closed areas, called here \"fundamental features\", the recognition was performed using a backpropagation neural network.","PeriodicalId":103287,"journal":{"name":"Proceedings II Workshop on Cybernetic Vision","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122011095","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 : 1996-12-09DOI: 10.1109/CYBVIS.1996.629433
M. von Grunau, M. Iordanova
In this paper, we examine some effects of the selection of spatial location on the processing of subsequent stimuli. Both bottom-up selection due to priming and top-down selection due to attention are discussed. In experiments with the motion induction illusion, we demonstrate a speed-up of processing due to selection, a center/surround organization of facilitation and inhibition, and the equivalence of priming and attention. More generally, it is suggested that selection effects can be understood as activation of locations in striate or extra-striate cortex and that they are governed by stimulus salience.
{"title":"Visual selection: facilitation due to stimulus saliency","authors":"M. von Grunau, M. Iordanova","doi":"10.1109/CYBVIS.1996.629433","DOIUrl":"https://doi.org/10.1109/CYBVIS.1996.629433","url":null,"abstract":"In this paper, we examine some effects of the selection of spatial location on the processing of subsequent stimuli. Both bottom-up selection due to priming and top-down selection due to attention are discussed. In experiments with the motion induction illusion, we demonstrate a speed-up of processing due to selection, a center/surround organization of facilitation and inhibition, and the equivalence of priming and attention. More generally, it is suggested that selection effects can be understood as activation of locations in striate or extra-striate cortex and that they are governed by stimulus salience.","PeriodicalId":103287,"journal":{"name":"Proceedings II Workshop on Cybernetic Vision","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122272423","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 : 1996-12-09DOI: 10.1109/CYBVIS.1996.629440
R. J. Machado, P. Neves
In this paper we describe a simple hybrid architecture of multi-model neural network aimed at enhancing the accuracy of classification in image interpretation problems. We adopt a modular architecture with one neural network dedicated to each class of the problem domain, allowing each of these neural modules to be built according to a different paradigm. The selection of the paradigm for each class is based on a benchmark among a set of competitor neural network models. We demonstrate experimentally the effectiveness of this approach in the problem of deforestation monitoring in the Amazon region.
{"title":"Multi-model neural network for image classification","authors":"R. J. Machado, P. Neves","doi":"10.1109/CYBVIS.1996.629440","DOIUrl":"https://doi.org/10.1109/CYBVIS.1996.629440","url":null,"abstract":"In this paper we describe a simple hybrid architecture of multi-model neural network aimed at enhancing the accuracy of classification in image interpretation problems. We adopt a modular architecture with one neural network dedicated to each class of the problem domain, allowing each of these neural modules to be built according to a different paradigm. The selection of the paradigm for each class is based on a benchmark among a set of competitor neural network models. We demonstrate experimentally the effectiveness of this approach in the problem of deforestation monitoring in the Amazon region.","PeriodicalId":103287,"journal":{"name":"Proceedings II Workshop on Cybernetic Vision","volume":"101 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116946587","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 : 1996-12-09DOI: 10.1109/CYBVIS.1996.629450
Y.R. Venturini, P. E. Cruvinel
This paper deals with the implementation of an algorithm for digital image quality analysis based on the Wiener technique. Image quality is evaluated through the analysis of image noise. The algorithm for Wiener spectral analysis is based on both Fourier transform and random process techniques. An interface with several menus for image analysis is also included. The results obtained with implementation of the algorithm indicate its potential for image quality analysis and its applicability not only to tomographic images but also to images obtained through other systems, which include both acquisition and digital reconstruction procedures.
{"title":"Analysis of the quality of digital images through a noise spectrum","authors":"Y.R. Venturini, P. E. Cruvinel","doi":"10.1109/CYBVIS.1996.629450","DOIUrl":"https://doi.org/10.1109/CYBVIS.1996.629450","url":null,"abstract":"This paper deals with the implementation of an algorithm for digital image quality analysis based on the Wiener technique. Image quality is evaluated through the analysis of image noise. The algorithm for Wiener spectral analysis is based on both Fourier transform and random process techniques. An interface with several menus for image analysis is also included. The results obtained with implementation of the algorithm indicate its potential for image quality analysis and its applicability not only to tomographic images but also to images obtained through other systems, which include both acquisition and digital reconstruction procedures.","PeriodicalId":103287,"journal":{"name":"Proceedings II Workshop on Cybernetic Vision","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117020344","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 : 1996-12-09DOI: 10.1109/CYBVIS.1996.629436
S. Neuenschwander, M. Castelo‐Branco, S. Herculano, W. Singer
Synchronous oscillations within a broad range of frequencies have been described for visual responses in retinal ganglion cells which are transmitted by the lateral geniculate nucleus (LGN), raising the question of how oscillatory inputs contribute to synchronous oscillatory responses in the cortex. We have made simultaneous recordings from visual areas 17 and 18 as well as the LGN and retina of the anaesthetized cat to examine how the occurrence of synchronization in the cortex follows subcortical oscillatory responses. Strong correlation of oscillatory responses was observed between the retina, LGN and the cortex. This finding provides evidence that cortical neurons may follow retinal oscillatory responses relayed by the LGN. In many cases, however, the development of cortical synchronization was independent of the temporal structure of the incoming inputs, excluding a simple feedforward mechanism for cortical synchronization. It is possible that depending on the stimulus, synchronous input from the LGN could facilitate synchronization of cortical responses by intracortical mechanisms and thereby contribute to the binding of features of objects.
{"title":"Synchronous oscillations in the cortex, LGN and retina of the cat: how are they related?","authors":"S. Neuenschwander, M. Castelo‐Branco, S. Herculano, W. Singer","doi":"10.1109/CYBVIS.1996.629436","DOIUrl":"https://doi.org/10.1109/CYBVIS.1996.629436","url":null,"abstract":"Synchronous oscillations within a broad range of frequencies have been described for visual responses in retinal ganglion cells which are transmitted by the lateral geniculate nucleus (LGN), raising the question of how oscillatory inputs contribute to synchronous oscillatory responses in the cortex. We have made simultaneous recordings from visual areas 17 and 18 as well as the LGN and retina of the anaesthetized cat to examine how the occurrence of synchronization in the cortex follows subcortical oscillatory responses. Strong correlation of oscillatory responses was observed between the retina, LGN and the cortex. This finding provides evidence that cortical neurons may follow retinal oscillatory responses relayed by the LGN. In many cases, however, the development of cortical synchronization was independent of the temporal structure of the incoming inputs, excluding a simple feedforward mechanism for cortical synchronization. It is possible that depending on the stimulus, synchronous input from the LGN could facilitate synchronization of cortical responses by intracortical mechanisms and thereby contribute to the binding of features of objects.","PeriodicalId":103287,"journal":{"name":"Proceedings II Workshop on Cybernetic Vision","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123039324","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 : 1996-12-09DOI: 10.1109/CYBVIS.1996.629444
C. Ménigault
Connectionist systems for spatio-temporal pattern recognition usually use a spatial coding of time or a mixed, spatial and temporal, representation. In this paper, we propose a temporal coding of spatial information by means of relative temporal delays of spatial events. These spatial events, translated into temporal events by scanning the 2D-space, feed a guided propagation network for handwritten records recognition. The network can adapt to (i) global pattern translation, (ii) local spatial shift of spatial events around their original position, (iii) suppression of some spatial events, and (iv) global pattern transformation.
{"title":"Temporal representation of space for adaptation to distortions among 2 dimensions","authors":"C. Ménigault","doi":"10.1109/CYBVIS.1996.629444","DOIUrl":"https://doi.org/10.1109/CYBVIS.1996.629444","url":null,"abstract":"Connectionist systems for spatio-temporal pattern recognition usually use a spatial coding of time or a mixed, spatial and temporal, representation. In this paper, we propose a temporal coding of spatial information by means of relative temporal delays of spatial events. These spatial events, translated into temporal events by scanning the 2D-space, feed a guided propagation network for handwritten records recognition. The network can adapt to (i) global pattern translation, (ii) local spatial shift of spatial events around their original position, (iii) suppression of some spatial events, and (iv) global pattern transformation.","PeriodicalId":103287,"journal":{"name":"Proceedings II Workshop on Cybernetic Vision","volume":"140 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116367543","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 : 1996-12-09DOI: 10.1109/CYBVIS.1996.629457
C. Galera, E.J. Lopes
We investigated the effect of noise density display organization and the number of stimuli in a visual search task. The results show that when the stimuli were randomly distributed in the visual field, the target was detected in parallel in intact displays and was serially searched in noisy displays; in this case search difficulty increases with noise density. When the display was organized, the search was always parallel. These results suggest that noise and display organization act on early visual processes which affect the size of attentional focus.
{"title":"Diagnostics of parallel and serial processing in a visual search task","authors":"C. Galera, E.J. Lopes","doi":"10.1109/CYBVIS.1996.629457","DOIUrl":"https://doi.org/10.1109/CYBVIS.1996.629457","url":null,"abstract":"We investigated the effect of noise density display organization and the number of stimuli in a visual search task. The results show that when the stimuli were randomly distributed in the visual field, the target was detected in parallel in intact displays and was serially searched in noisy displays; in this case search difficulty increases with noise density. When the display was organized, the search was always parallel. These results suggest that noise and display organization act on early visual processes which affect the size of attentional focus.","PeriodicalId":103287,"journal":{"name":"Proceedings II Workshop on Cybernetic Vision","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115310158","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：