Pub Date : 2002-11-07DOI: 10.1109/ICDSP.2002.1028307
M. Martínez‐Ramón, J. Arenas-García, Á. Navia-Vázquez, A. Figueiras-Vidal
In this paper, we propose to adaptively combine two LMS adaptive transversal filters for plant identification. One of the filters has a high and the other a low adaption step, in order to combine good tracking capabilities under (fast) change conditions with a reduced convergence error along stationary periods. A brief discussion of the characteristics of the combination is included, emphasizing that it allows the possibility of dealing with "intermediate" rate of change situations, in opposition to (implicit or explicit) switching mechanisms. A selected illustrative simulation example shows the effectiveness of this approach. Some complementary lines of research are indicated, from the points of view of improving the algorithm and of extending the fields of application.
{"title":"An adaptive combination of adaptive filters for plant identification","authors":"M. Martínez‐Ramón, J. Arenas-García, Á. Navia-Vázquez, A. Figueiras-Vidal","doi":"10.1109/ICDSP.2002.1028307","DOIUrl":"https://doi.org/10.1109/ICDSP.2002.1028307","url":null,"abstract":"In this paper, we propose to adaptively combine two LMS adaptive transversal filters for plant identification. One of the filters has a high and the other a low adaption step, in order to combine good tracking capabilities under (fast) change conditions with a reduced convergence error along stationary periods. A brief discussion of the characteristics of the combination is included, emphasizing that it allows the possibility of dealing with \"intermediate\" rate of change situations, in opposition to (implicit or explicit) switching mechanisms. A selected illustrative simulation example shows the effectiveness of this approach. Some complementary lines of research are indicated, from the points of view of improving the algorithm and of extending the fields of application.","PeriodicalId":351073,"journal":{"name":"2002 14th International Conference on Digital Signal Processing Proceedings. DSP 2002 (Cat. No.02TH8628)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121630749","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 : 2002-11-07DOI: 10.1109/ICDSP.2002.1028249
A. Krot
Decomposition methods of nonlinear operators describing the behavior of system in state space (phase space) are very important for analysis, identification and modeling of nonlinear dynamical systems (NDS), in particular NDS with self-organization (or complex NDS). The aim of this paper is derivation and classification of matrix series describing decomposition of vector functions from phase space variables and NDS operators into state space. This paper also develops some statements of matrix decomposition and main principles for analysis of attractors of complex NDS.
{"title":"Application of expansion into matrix series to analysis of attractors of complex nonlinear dynamical systems","authors":"A. Krot","doi":"10.1109/ICDSP.2002.1028249","DOIUrl":"https://doi.org/10.1109/ICDSP.2002.1028249","url":null,"abstract":"Decomposition methods of nonlinear operators describing the behavior of system in state space (phase space) are very important for analysis, identification and modeling of nonlinear dynamical systems (NDS), in particular NDS with self-organization (or complex NDS). The aim of this paper is derivation and classification of matrix series describing decomposition of vector functions from phase space variables and NDS operators into state space. This paper also develops some statements of matrix decomposition and main principles for analysis of attractors of complex NDS.","PeriodicalId":351073,"journal":{"name":"2002 14th International Conference on Digital Signal Processing Proceedings. DSP 2002 (Cat. No.02TH8628)","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121187926","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 : 2002-11-07DOI: 10.1109/ICDSP.2002.1028261
Yuhua Ding, G. Vachtsevanos, A. Yezzi, Yinchuan Zhang, Y. Wardi
The segmentation accuracy is shown to be a critical factor in detection rate improvement. With accurate segmentation, results are easier to interpret, and classification performance is better. Therefore, it is required to have a performance measure for segmentation evaluation. However, a number of restrictions limit using existing segmentation performance measures. A recursive segmentation and classification scheme is proposed to improve segmentation accuracy and classification performance in real-time machine vision applications. In this scheme, the confidence level of classification results is used as a new performance measure to evaluate the accuracy of segmentation algorithm. Segmentation is repeated until a classification with desired confidence level is achieved. This scheme can be implemented automatically. Experimental results show that it is efficient to improve segmentation accuracy and the overall detection performance, especially for real-time machine vision applications, where the scene is complicated and a single segmentation algorithm cannot produce satisfactory results.
{"title":"A recursive segmentation and classification scheme for improving segmentation accuracy and detection rate in real-time machine vision applications","authors":"Yuhua Ding, G. Vachtsevanos, A. Yezzi, Yinchuan Zhang, Y. Wardi","doi":"10.1109/ICDSP.2002.1028261","DOIUrl":"https://doi.org/10.1109/ICDSP.2002.1028261","url":null,"abstract":"The segmentation accuracy is shown to be a critical factor in detection rate improvement. With accurate segmentation, results are easier to interpret, and classification performance is better. Therefore, it is required to have a performance measure for segmentation evaluation. However, a number of restrictions limit using existing segmentation performance measures. A recursive segmentation and classification scheme is proposed to improve segmentation accuracy and classification performance in real-time machine vision applications. In this scheme, the confidence level of classification results is used as a new performance measure to evaluate the accuracy of segmentation algorithm. Segmentation is repeated until a classification with desired confidence level is achieved. This scheme can be implemented automatically. Experimental results show that it is efficient to improve segmentation accuracy and the overall detection performance, especially for real-time machine vision applications, where the scene is complicated and a single segmentation algorithm cannot produce satisfactory results.","PeriodicalId":351073,"journal":{"name":"2002 14th International Conference on Digital Signal Processing Proceedings. DSP 2002 (Cat. No.02TH8628)","volume":"81 3-4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131641189","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 : 2002-11-07DOI: 10.1109/ICDSP.2002.1028176
J. Bunch, R. L. Borne, I. Proudler
Along with its many desirable properties the fast transversal filter (FTF) algorithm suffers from explosive divergence. This type of divergence occurs when the algorithm is seemingly performing its operations normally, producing usable solutions, when the algorithm appears to suddenly produce extremely large errors and an obviously useless solution. Although it is known that a loss of backward consistency is the cause for the resultant perturbations, i.e., a violation to interrelationships between update parameters are not explicitly enforced by the update equations, it is not known why the algorithm suffers explosive divergence rather than a divergence that grows as a continuous function over time. Algorithms have been proposed to circumvent this problem but it remains to be shown through theoretical justification whether these algorithms have remedied the problem or only put it off to some later iteration. Here, we provide a rationale to explain the explosive character of divergence that is inherent to the manner in which the FTF algorithm is derived.
{"title":"Understanding the explosive divergence of the FTF algorithm","authors":"J. Bunch, R. L. Borne, I. Proudler","doi":"10.1109/ICDSP.2002.1028176","DOIUrl":"https://doi.org/10.1109/ICDSP.2002.1028176","url":null,"abstract":"Along with its many desirable properties the fast transversal filter (FTF) algorithm suffers from explosive divergence. This type of divergence occurs when the algorithm is seemingly performing its operations normally, producing usable solutions, when the algorithm appears to suddenly produce extremely large errors and an obviously useless solution. Although it is known that a loss of backward consistency is the cause for the resultant perturbations, i.e., a violation to interrelationships between update parameters are not explicitly enforced by the update equations, it is not known why the algorithm suffers explosive divergence rather than a divergence that grows as a continuous function over time. Algorithms have been proposed to circumvent this problem but it remains to be shown through theoretical justification whether these algorithms have remedied the problem or only put it off to some later iteration. Here, we provide a rationale to explain the explosive character of divergence that is inherent to the manner in which the FTF algorithm is derived.","PeriodicalId":351073,"journal":{"name":"2002 14th International Conference on Digital Signal Processing Proceedings. DSP 2002 (Cat. No.02TH8628)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133512210","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 : 2002-11-07DOI: 10.1109/ICDSP.2002.1028188
C. Hsieh, Pei-Ying Sou
The synchronization problem is a serious issue in the watermark area. It happens when losing the location of the embedded watermark when cropping, shifting and so on. Therefore the detector finds it difficult to reform the copyright information. In this paper, the proposed method combines an energy-feature basis idea in the time domain to solve the synchronization problem and achieved blind audio watermarking in the cepstrum domain. The simulation results show a high security performance against the MP3 attack and the robustness improvement of several kinds of digital distortion attacks such as pitch-shifting, and cut samples.
{"title":"Blind cepstrum domain audio watermarking based on time energy features","authors":"C. Hsieh, Pei-Ying Sou","doi":"10.1109/ICDSP.2002.1028188","DOIUrl":"https://doi.org/10.1109/ICDSP.2002.1028188","url":null,"abstract":"The synchronization problem is a serious issue in the watermark area. It happens when losing the location of the embedded watermark when cropping, shifting and so on. Therefore the detector finds it difficult to reform the copyright information. In this paper, the proposed method combines an energy-feature basis idea in the time domain to solve the synchronization problem and achieved blind audio watermarking in the cepstrum domain. The simulation results show a high security performance against the MP3 attack and the robustness improvement of several kinds of digital distortion attacks such as pitch-shifting, and cut samples.","PeriodicalId":351073,"journal":{"name":"2002 14th International Conference on Digital Signal Processing Proceedings. DSP 2002 (Cat. No.02TH8628)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131819282","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 : 2002-11-07DOI: 10.1109/ICDSP.2002.1028218
E. Jung, A. Schwarzbacher, R. Lawlor
Traditionally the interest in voice gender conversion was of a more theoretical nature rather than founded in real-life applications. However, with the increase in mobile communication and the resulting limitation in transmission bandwidth new approaches to minimising data rates have to be developed. Here voice gender normalisation (VGN) presents a novel method of achieving higher compression rates by using the VGN algorithm to remove all gender specific components of a speech signal and thus leaving only the information content to be transmitted. A second application for VGN is in the field of speech controlled systems, where current speech recognition algorithms have to deal with the voice characteristics of a speaker as well as the information content. Here again the use of VGN can remove the speakers voice characteristics leaving only the pure information. Therefore, such a system would be capable of achieving much higher recognition rates while being independent of the speaker. This paper presents the theory of a gender removal system based on VGN and furthermore, outlines an efficient real-time hardware implementation for use in portable communications equipment.
{"title":"Implementation of real-time AMDF pitch-detection for voice gender normalisation","authors":"E. Jung, A. Schwarzbacher, R. Lawlor","doi":"10.1109/ICDSP.2002.1028218","DOIUrl":"https://doi.org/10.1109/ICDSP.2002.1028218","url":null,"abstract":"Traditionally the interest in voice gender conversion was of a more theoretical nature rather than founded in real-life applications. However, with the increase in mobile communication and the resulting limitation in transmission bandwidth new approaches to minimising data rates have to be developed. Here voice gender normalisation (VGN) presents a novel method of achieving higher compression rates by using the VGN algorithm to remove all gender specific components of a speech signal and thus leaving only the information content to be transmitted. A second application for VGN is in the field of speech controlled systems, where current speech recognition algorithms have to deal with the voice characteristics of a speaker as well as the information content. Here again the use of VGN can remove the speakers voice characteristics leaving only the pure information. Therefore, such a system would be capable of achieving much higher recognition rates while being independent of the speaker. This paper presents the theory of a gender removal system based on VGN and furthermore, outlines an efficient real-time hardware implementation for use in portable communications equipment.","PeriodicalId":351073,"journal":{"name":"2002 14th International Conference on Digital Signal Processing Proceedings. DSP 2002 (Cat. No.02TH8628)","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114151640","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 : 2002-11-07DOI: 10.1109/ICDSP.2002.1028212
S. Billings, G. Newsam
Sharpening and edge-enhancement filters are often applied to geophysical data that were collected with an uneven sample spacing and varying sample density. In this paper we propose an alternative methodology to the usual gridding/digital filtering processing pathway. We first fit a continuous global surface (CGS) to the data and then implicitly apply Fourier domain filtering to the entire surface. The CGS is constructed to optimize some property of the surface (e.g. smoothness). We find that the best approach is to optimize the properties of the filtered surface rather than the surface that fits the data. Otherwise certain filters cause the transformed surface to have singularities at the data points. We demonstrate the viability of the methodology in the computation of the second vertical derivative of a gravity survey.
{"title":"Fourier filtering of continuous global surfaces","authors":"S. Billings, G. Newsam","doi":"10.1109/ICDSP.2002.1028212","DOIUrl":"https://doi.org/10.1109/ICDSP.2002.1028212","url":null,"abstract":"Sharpening and edge-enhancement filters are often applied to geophysical data that were collected with an uneven sample spacing and varying sample density. In this paper we propose an alternative methodology to the usual gridding/digital filtering processing pathway. We first fit a continuous global surface (CGS) to the data and then implicitly apply Fourier domain filtering to the entire surface. The CGS is constructed to optimize some property of the surface (e.g. smoothness). We find that the best approach is to optimize the properties of the filtered surface rather than the surface that fits the data. Otherwise certain filters cause the transformed surface to have singularities at the data points. We demonstrate the viability of the methodology in the computation of the second vertical derivative of a gravity survey.","PeriodicalId":351073,"journal":{"name":"2002 14th International Conference on Digital Signal Processing Proceedings. DSP 2002 (Cat. No.02TH8628)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114164424","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 : 2002-11-07DOI: 10.1109/ICDSP.2002.1028191
N. Thomos, N. Boulgouris, E. Kokkinou, M. Strintzis
In this paper, the sequential decoding of convolutional codes is proposed for data hiding in the wavelet domain. The performance of this technique is evaluated for data hiding in JPEG2000 images and is shown to be advantageous in comparison to other methods for the embedding/extraction of information in digital images.
{"title":"Efficient data hiding in JPEG2000 images using sequential decoding of convolutional codes","authors":"N. Thomos, N. Boulgouris, E. Kokkinou, M. Strintzis","doi":"10.1109/ICDSP.2002.1028191","DOIUrl":"https://doi.org/10.1109/ICDSP.2002.1028191","url":null,"abstract":"In this paper, the sequential decoding of convolutional codes is proposed for data hiding in the wavelet domain. The performance of this technique is evaluated for data hiding in JPEG2000 images and is shown to be advantageous in comparison to other methods for the embedding/extraction of information in digital images.","PeriodicalId":351073,"journal":{"name":"2002 14th International Conference on Digital Signal Processing Proceedings. DSP 2002 (Cat. No.02TH8628)","volume":"756 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116117427","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 : 2002-11-07DOI: 10.1109/ICDSP.2002.1028144
D. Sabino, E. K. Nakamura, L. Costa, R. Calado, M. Zago
A new approach to chromatin texture characterization is proposed, based on the recently introduced concept of multiscale fractal dimension. Promising results for differentiating normal from abnormal blood cells have been obtained by considering the peaks of multiscale fractal dimension after Minkowski-Bouligand dilation of the nucleus gray tones. Results for a lymphocyte database exemplify the potential of the method with respect to nuclear texture discrimination. A brief review of related works is also included, focusing on statistical approaches.
{"title":"Chromatin texture characterization using multiscale fractal dimension","authors":"D. Sabino, E. K. Nakamura, L. Costa, R. Calado, M. Zago","doi":"10.1109/ICDSP.2002.1028144","DOIUrl":"https://doi.org/10.1109/ICDSP.2002.1028144","url":null,"abstract":"A new approach to chromatin texture characterization is proposed, based on the recently introduced concept of multiscale fractal dimension. Promising results for differentiating normal from abnormal blood cells have been obtained by considering the peaks of multiscale fractal dimension after Minkowski-Bouligand dilation of the nucleus gray tones. Results for a lymphocyte database exemplify the potential of the method with respect to nuclear texture discrimination. A brief review of related works is also included, focusing on statistical approaches.","PeriodicalId":351073,"journal":{"name":"2002 14th International Conference on Digital Signal Processing Proceedings. DSP 2002 (Cat. No.02TH8628)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115158174","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 : 2002-11-07DOI: 10.1109/ICDSP.2002.1028165
E. Dellandréa, P. Makris, N. Vincent, M. Boiron
We present in this paper a medical signal analysis method that we have developed in order to study, in an atraumatic and non-invasive way, the inferior oesophageal sphincter behaviour. This one plays an important part in the gastro-oesophageal reflux phenomenon. The proposed method is more efficient and is better suited to our problem than traditional signal processing methods that did not give good results. It has a double originality. The first one lies on the audio signal coding, based on its time-frequency representation, and allowing the extraction of a set of ordered words that makes up a text. The second one lies on the use of the Zipf law as an analysis tool. The parameters obtained from the Zipf law observation allow one to automatically detect and characterize signal regions of interest, which are periods corresponding to xiphoidal sounds (produced when a bolus, a small quantity of baryte, crosses the inferior oesophageal sphincter). Thus, the results permit one to evaluate the sphincter capability, which is a precious help for specialists.
{"title":"A medical acoustic signal analysis method based on Zipf law","authors":"E. Dellandréa, P. Makris, N. Vincent, M. Boiron","doi":"10.1109/ICDSP.2002.1028165","DOIUrl":"https://doi.org/10.1109/ICDSP.2002.1028165","url":null,"abstract":"We present in this paper a medical signal analysis method that we have developed in order to study, in an atraumatic and non-invasive way, the inferior oesophageal sphincter behaviour. This one plays an important part in the gastro-oesophageal reflux phenomenon. The proposed method is more efficient and is better suited to our problem than traditional signal processing methods that did not give good results. It has a double originality. The first one lies on the audio signal coding, based on its time-frequency representation, and allowing the extraction of a set of ordered words that makes up a text. The second one lies on the use of the Zipf law as an analysis tool. The parameters obtained from the Zipf law observation allow one to automatically detect and characterize signal regions of interest, which are periods corresponding to xiphoidal sounds (produced when a bolus, a small quantity of baryte, crosses the inferior oesophageal sphincter). Thus, the results permit one to evaluate the sphincter capability, which is a precious help for specialists.","PeriodicalId":351073,"journal":{"name":"2002 14th International Conference on Digital Signal Processing Proceedings. DSP 2002 (Cat. No.02TH8628)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123436486","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：