Stack filters arc a class of nonlinear filters, first introduced by Wedcnt et. al. Stack filters perform well in many situations where linear filters fail. Stack fillers include rank order fillers, morphological fillers and weighted median fillers. The stack filter is defined by a Boolean function. The output of Boolean functions is restricted two values (i.e., "0" or "1"), Intuitively, one would expect better performance for stack filters, if the output of Boolean functions is defined from 0 to 1 continuously. We call this Boolean functions fuzzy Boolean functions. We discuss about fuzzy center weighted median (FCWM) filters which is one of the simplest fuzzy stack filters in this paper. Two design methods are shown in this paper.
{"title":"Fuzzy center weighted median filters","authors":"A. Taguchi, N. Izawa","doi":"10.5281/ZENODO.36430","DOIUrl":"https://doi.org/10.5281/ZENODO.36430","url":null,"abstract":"Stack filters arc a class of nonlinear filters, first introduced by Wedcnt et. al. Stack filters perform well in many situations where linear filters fail. Stack fillers include rank order fillers, morphological fillers and weighted median fillers. The stack filter is defined by a Boolean function. The output of Boolean functions is restricted two values (i.e., \"0\" or \"1\"), Intuitively, one would expect better performance for stack filters, if the output of Boolean functions is defined from 0 to 1 continuously. We call this Boolean functions fuzzy Boolean functions. We discuss about fuzzy center weighted median (FCWM) filters which is one of the simplest fuzzy stack filters in this paper. Two design methods are shown in this paper.","PeriodicalId":282153,"journal":{"name":"1996 8th European Signal Processing Conference (EUSIPCO 1996)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116069581","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}
We present the use of a cross-correlated quantization noise model in the recently proposed Kalman innovation speech coding scheme. Computer simulations and informal listening tests indicate that the incorporation of a cross-correlated noise model yields an improvement in both SNR and perceptual quality when compared to a uncorrelated noise model.
{"title":"Innovation coding with a cross-correlated quantization noise model","authors":"S. Andersen, M. Olesen, S. H. Jensen, E. Hansen","doi":"10.5281/ZENODO.36286","DOIUrl":"https://doi.org/10.5281/ZENODO.36286","url":null,"abstract":"We present the use of a cross-correlated quantization noise model in the recently proposed Kalman innovation speech coding scheme. Computer simulations and informal listening tests indicate that the incorporation of a cross-correlated noise model yields an improvement in both SNR and perceptual quality when compared to a uncorrelated noise model.","PeriodicalId":282153,"journal":{"name":"1996 8th European Signal Processing Conference (EUSIPCO 1996)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125643067","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}
In this paper, we present a new linear modulation classification method based on a fourth-order cumulant of the stationary signal. Under some hypothesis, this method can be applied to carrier-modulated or baseband signals and doesn't need the knowledge of the signal to noise ratio. An example of classification is given for 4 PSK vs. 16 QAM modulations. Theoretical performance are approximated and compared to simulation results. The system achieves more than 90 % of correct classification for only 500 transmitted symbols and a signal to noise ratio of 0 dB.
{"title":"Classification of linear modulations by mean of a fourth-order cumulant","authors":"Denys Boiteau, C. Martret","doi":"10.5281/ZENODO.36181","DOIUrl":"https://doi.org/10.5281/ZENODO.36181","url":null,"abstract":"In this paper, we present a new linear modulation classification method based on a fourth-order cumulant of the stationary signal. Under some hypothesis, this method can be applied to carrier-modulated or baseband signals and doesn't need the knowledge of the signal to noise ratio. An example of classification is given for 4 PSK vs. 16 QAM modulations. Theoretical performance are approximated and compared to simulation results. The system achieves more than 90 % of correct classification for only 500 transmitted symbols and a signal to noise ratio of 0 dB.","PeriodicalId":282153,"journal":{"name":"1996 8th European Signal Processing Conference (EUSIPCO 1996)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125568033","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}
Reconstruction of 3D shape of the solder paste printed on SMT component pads is a major inspection task in the PCB manufacturing process. The paper reports on the use of phase profilometry for this inspection task. In phase profilometry a structured light pattern is projected onto the object and viewed by a camera. Since the imaged pattern is phase-modulated according to the topography of the object, the extraction of phase information from the image enables reconstructing the 3D shape. In this paper two phase-extraction methods, Fourier Transform Profilometry and Signal Domain Profilometry, are compared by means of simulations and experiments. Results show that the Fourier method performs better, yielding neat detection of the elevation with respect to PCB surface associated with solder paste.
{"title":"Three-dimensional inspection of printed circuit boards using phase profilometry","authors":"L. D. Stefano, F. Boland","doi":"10.5281/ZENODO.36098","DOIUrl":"https://doi.org/10.5281/ZENODO.36098","url":null,"abstract":"Reconstruction of 3D shape of the solder paste printed on SMT component pads is a major inspection task in the PCB manufacturing process. The paper reports on the use of phase profilometry for this inspection task. In phase profilometry a structured light pattern is projected onto the object and viewed by a camera. Since the imaged pattern is phase-modulated according to the topography of the object, the extraction of phase information from the image enables reconstructing the 3D shape. In this paper two phase-extraction methods, Fourier Transform Profilometry and Signal Domain Profilometry, are compared by means of simulations and experiments. Results show that the Fourier method performs better, yielding neat detection of the elevation with respect to PCB surface associated with solder paste.","PeriodicalId":282153,"journal":{"name":"1996 8th European Signal Processing Conference (EUSIPCO 1996)","volume":"142 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121003405","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}
P. Sirotti, Mauro Zanchetti, G. Rizzatto, F. Stacul
The recognition of echotexture in echographic images may fail due to the distortions introduced by the scan system. We have implemented a rotation and scale invariant recognition method of echographic textures. The significant features assumed to characterise the images are vectors whose components are the values of a modified Fourier transform (MFT) of the images. Our method assures a good reliability and allows a short computation time, also when implemented on small computers. The method has till now been proved over breast and thyroid images, exhibiting a very good discrimination capability.
{"title":"Modified fourier transform recognition of echographic images","authors":"P. Sirotti, Mauro Zanchetti, G. Rizzatto, F. Stacul","doi":"10.5281/ZENODO.36162","DOIUrl":"https://doi.org/10.5281/ZENODO.36162","url":null,"abstract":"The recognition of echotexture in echographic images may fail due to the distortions introduced by the scan system. We have implemented a rotation and scale invariant recognition method of echographic textures. The significant features assumed to characterise the images are vectors whose components are the values of a modified Fourier transform (MFT) of the images. Our method assures a good reliability and allows a short computation time, also when implemented on small computers. The method has till now been proved over breast and thyroid images, exhibiting a very good discrimination capability.","PeriodicalId":282153,"journal":{"name":"1996 8th European Signal Processing Conference (EUSIPCO 1996)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121035378","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}
Linear Discriminant Analysis (LDA) has been applied successfully to speech recognition tasks, improving accuracy and robustness against some types of noise. However, it is well known that LDA suffers from some weaknesses if the distributions are not unimodal or when the mean of the distributions are shared. In this paper, we propose to take advantage of the nonlinear discriminant properties of the Artificial Neural Networks (ANN) in the task of reducing the dimensionality of the input space, leading to a nonlinear discriminant analysis.
{"title":"Nonlinear discriminant analysis with neural networks for speech recognition","authors":"V. Fontaine, C. Ris, H. Leich","doi":"10.5281/ZENODO.36303","DOIUrl":"https://doi.org/10.5281/ZENODO.36303","url":null,"abstract":"Linear Discriminant Analysis (LDA) has been applied successfully to speech recognition tasks, improving accuracy and robustness against some types of noise. However, it is well known that LDA suffers from some weaknesses if the distributions are not unimodal or when the mean of the distributions are shared. In this paper, we propose to take advantage of the nonlinear discriminant properties of the Artificial Neural Networks (ANN) in the task of reducing the dimensionality of the input space, leading to a nonlinear discriminant analysis.","PeriodicalId":282153,"journal":{"name":"1996 8th European Signal Processing Conference (EUSIPCO 1996)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124855259","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}
In this paper, a new 2-D block LMS algorithm is presented. This algorithm, which is an exact mathematical formulation of classical 2-D LMS algorithms, presents the advantage of preserving a good convergence as the block size increases. The reduction in the computational complexity is achieved by exploiting the redundancy between successive computations, rather than using disjoint or partially overlapping windows. The latter are known to degrade the convergence when the block size is large.
{"title":"A new two-dimensional block least mean squares adaptive algorithm","authors":"S. Attallah, M. Najim","doi":"10.5281/ZENODO.36070","DOIUrl":"https://doi.org/10.5281/ZENODO.36070","url":null,"abstract":"In this paper, a new 2-D block LMS algorithm is presented. This algorithm, which is an exact mathematical formulation of classical 2-D LMS algorithms, presents the advantage of preserving a good convergence as the block size increases. The reduction in the computational complexity is achieved by exploiting the redundancy between successive computations, rather than using disjoint or partially overlapping windows. The latter are known to degrade the convergence when the block size is large.","PeriodicalId":282153,"journal":{"name":"1996 8th European Signal Processing Conference (EUSIPCO 1996)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115086280","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}
D. A. Valkaniotis, J. Sirigos, N. Fakotakis, G. Kokkinakis
In this paper we present a text-independent offline writer recognition system based on multilayer perceptrons (MLPs). The system can be used for both identification and verification purposes. It was tested on a population of 20 writers with non-correlated training and test specimens. The mean error for identification was 3.5% while error rates as low as 0.5% were achieved on specimens with more than 25 characters. For verification the mean error was 1.2% (2.22% false rejection, 0.18% false acceptance) considering a minimum of 15 characters per test specimen. These error rates are comparable to those achieved by classical methods while the response of the system is substantially faster.
{"title":"Text-independent off-line writer recognition using neural networks","authors":"D. A. Valkaniotis, J. Sirigos, N. Fakotakis, G. Kokkinakis","doi":"10.5281/ZENODO.36306","DOIUrl":"https://doi.org/10.5281/ZENODO.36306","url":null,"abstract":"In this paper we present a text-independent offline writer recognition system based on multilayer perceptrons (MLPs). The system can be used for both identification and verification purposes. It was tested on a population of 20 writers with non-correlated training and test specimens. The mean error for identification was 3.5% while error rates as low as 0.5% were achieved on specimens with more than 25 characters. For verification the mean error was 1.2% (2.22% false rejection, 0.18% false acceptance) considering a minimum of 15 characters per test specimen. These error rates are comparable to those achieved by classical methods while the response of the system is substantially faster.","PeriodicalId":282153,"journal":{"name":"1996 8th European Signal Processing Conference (EUSIPCO 1996)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122729498","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}
Perfect reconstruction FIR filter banks implement a particular class of signal expansions in ℓ2(Z). These expansions are studied in this paper. Necessary and sufficient conditions on an FIR filter bank to implement a frame or a tight frame decomposition are given, as well as the necessary and sufficient condition for a feasibility of perfect reconstruction using FIR filters. Complete parameterizations of FIR filter banks satisfying these conditions are given. Further, we study the condition under which the minimal dual frame to the frame associated to an FIR filter bank is also FIR, and give a parameterization of a class of filter banks having this property. We then concentrate on the least constrained class, namely nonsubsampled filter banks, for which these frame conditions have particular forms.
{"title":"FIR oversampled filter banks and frames in ℓ2(Z)","authors":"Z. Cvetković, M. Vetterli","doi":"10.5281/ZENODO.36403","DOIUrl":"https://doi.org/10.5281/ZENODO.36403","url":null,"abstract":"Perfect reconstruction FIR filter banks implement a particular class of signal expansions in ℓ2(Z). These expansions are studied in this paper. Necessary and sufficient conditions on an FIR filter bank to implement a frame or a tight frame decomposition are given, as well as the necessary and sufficient condition for a feasibility of perfect reconstruction using FIR filters. Complete parameterizations of FIR filter banks satisfying these conditions are given. Further, we study the condition under which the minimal dual frame to the frame associated to an FIR filter bank is also FIR, and give a parameterization of a class of filter banks having this property. We then concentrate on the least constrained class, namely nonsubsampled filter banks, for which these frame conditions have particular forms.","PeriodicalId":282153,"journal":{"name":"1996 8th European Signal Processing Conference (EUSIPCO 1996)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114603065","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}
Durbin's method for Moving Average (MA) estimation uses the estimated parameters of a long AutoRegressive (AR) model to compute the desired MA parameters. A theoretical order for that long AR model is ∞, but very high AR orders lead to inaccurate MA models in the finite sample practice. A new theoretical argument is presented to derive an expression for the best finite long AR order for a known MA process and a given sample size. Intermediate AR models of precisely that order produce the most accurate MA models. This new order differs from the best AR order to be used for prediction. An algorithm is presented that enables use of the theory for the best long AR order in known processes to data of an unknown process.
{"title":"The best order of long autoregressive models for moving average estimation","authors":"P. Broersen","doi":"10.5281/ZENODO.35981","DOIUrl":"https://doi.org/10.5281/ZENODO.35981","url":null,"abstract":"Durbin's method for Moving Average (MA) estimation uses the estimated parameters of a long AutoRegressive (AR) model to compute the desired MA parameters. A theoretical order for that long AR model is ∞, but very high AR orders lead to inaccurate MA models in the finite sample practice. A new theoretical argument is presented to derive an expression for the best finite long AR order for a known MA process and a given sample size. Intermediate AR models of precisely that order produce the most accurate MA models. This new order differs from the best AR order to be used for prediction. An algorithm is presented that enables use of the theory for the best long AR order in known processes to data of an unknown process.","PeriodicalId":282153,"journal":{"name":"1996 8th European Signal Processing Conference (EUSIPCO 1996)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122151280","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：