Pub Date : 2012-07-02DOI: 10.1109/ISSPA.2012.6310630
William C. Suski, Salil Banerjee, A. Hoover
Previous works in ultra-wideband (UWB) noise modeling have mostly focused on isolating the individual sources of error. However, it is important to recognize that some errors will always pass through to the system output. In this work, we methodically evaluated the system-level noise of a UWB position tracking system. We define system-level noise as the measurement error obtained when the system is installed in a real-world environment. Our results show that a multi-modal noise model will be essential for filtering system-level noise. To encourage further research, all of our data has been made publicly available.
{"title":"System-level noise of an ultra-wideband tracking system","authors":"William C. Suski, Salil Banerjee, A. Hoover","doi":"10.1109/ISSPA.2012.6310630","DOIUrl":"https://doi.org/10.1109/ISSPA.2012.6310630","url":null,"abstract":"Previous works in ultra-wideband (UWB) noise modeling have mostly focused on isolating the individual sources of error. However, it is important to recognize that some errors will always pass through to the system output. In this work, we methodically evaluated the system-level noise of a UWB position tracking system. We define system-level noise as the measurement error obtained when the system is installed in a real-world environment. Our results show that a multi-modal noise model will be essential for filtering system-level noise. To encourage further research, all of our data has been made publicly available.","PeriodicalId":248763,"journal":{"name":"2012 11th International Conference on Information Science, Signal Processing and their Applications (ISSPA)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121235039","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 : 2012-07-02DOI: 10.1109/ISSPA.2012.6310688
Jalil Jalili, H. Rabbani, M. Akhlaghi, R. Kafieh, A. M. Dehnavi
Optical coherence tomography (OCT) is an effective and noninvasive modality for retinal imaging. 3-D data that acquired from 3-D Spectral Domain OCT (SD-OCT) have shown their importance in the evaluation of retinal diseases. In addition, this set of data provides an opportunity to study depth of retina. In this paper, we focus on forming X-Y axis images from each layer of retina. In this manner, we first choose diffusion map based segmentation for localization of 12 different boundaries in 3D retinal data. Then we take an average on layers which located between each pairs of detected boundaries. Therefore, we make the X-Y axis image from each layer. With wavelet based image fusion, we combine together the layers with appropriate information to make images with additional information in retinal depth.
{"title":"Forming projection images from each layer of retina using diffusion may based OCT segmentation","authors":"Jalil Jalili, H. Rabbani, M. Akhlaghi, R. Kafieh, A. M. Dehnavi","doi":"10.1109/ISSPA.2012.6310688","DOIUrl":"https://doi.org/10.1109/ISSPA.2012.6310688","url":null,"abstract":"Optical coherence tomography (OCT) is an effective and noninvasive modality for retinal imaging. 3-D data that acquired from 3-D Spectral Domain OCT (SD-OCT) have shown their importance in the evaluation of retinal diseases. In addition, this set of data provides an opportunity to study depth of retina. In this paper, we focus on forming X-Y axis images from each layer of retina. In this manner, we first choose diffusion map based segmentation for localization of 12 different boundaries in 3D retinal data. Then we take an average on layers which located between each pairs of detected boundaries. Therefore, we make the X-Y axis image from each layer. With wavelet based image fusion, we combine together the layers with appropriate information to make images with additional information in retinal depth.","PeriodicalId":248763,"journal":{"name":"2012 11th International Conference on Information Science, Signal Processing and their Applications (ISSPA)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125334191","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 : 2012-07-02DOI: 10.1109/ISSPA.2012.6310497
Ayman Rabee, I. Barhumi
In this paper we propose a highly reliable ECG analysis and classification approach using discrete wavelet transform multiresolution analysis and support vector machine (SVM). This approach is composed of three stages, including ECG signal preprocessing, feature selection, and classification of ECG beats. Wavelet transform is used for signal preprocessing, denoising, and for extracting the coefficients of the transform as features of each ECG beat which are employed as inputs to the classifier. SVM is used to construct a classifier to categorize the input ECG beat into one of 14 classes. In this work, 17260 ECG beats, including 14 different beat types, were selected from the MIT/BIH arrhythmia database. The average accuracy of classification for recognition of the 14 heart beat types is 99.2%.
{"title":"ECG signal classification using support vector machine based on wavelet multiresolution analysis","authors":"Ayman Rabee, I. Barhumi","doi":"10.1109/ISSPA.2012.6310497","DOIUrl":"https://doi.org/10.1109/ISSPA.2012.6310497","url":null,"abstract":"In this paper we propose a highly reliable ECG analysis and classification approach using discrete wavelet transform multiresolution analysis and support vector machine (SVM). This approach is composed of three stages, including ECG signal preprocessing, feature selection, and classification of ECG beats. Wavelet transform is used for signal preprocessing, denoising, and for extracting the coefficients of the transform as features of each ECG beat which are employed as inputs to the classifier. SVM is used to construct a classifier to categorize the input ECG beat into one of 14 classes. In this work, 17260 ECG beats, including 14 different beat types, were selected from the MIT/BIH arrhythmia database. The average accuracy of classification for recognition of the 14 heart beat types is 99.2%.","PeriodicalId":248763,"journal":{"name":"2012 11th International Conference on Information Science, Signal Processing and their Applications (ISSPA)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121705350","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 : 2012-07-02DOI: 10.1109/ISSPA.2012.6310623
D. Dubois, R. Lepage
Recent disasters have shown that there is a growing interest for remotely sensed data to support decision makers and emergency teams in the field. Fast and accurate detection of buildings and sustained damage is of great importance. Current methods rely on numerous photo-interpreters to visually analyze the data. Multiple pixel-based methods exist to classify pixels as being part of a building or not but results vary widely and precision is often poor with very high resolution images. This paper proposes an object-based solution to building detection and compares it to a traditional approach. Object-based classification clearly provides adequate results in much less time and thus is ideal for disaster response.
{"title":"Object- versus pixel-based building detection for disaster response","authors":"D. Dubois, R. Lepage","doi":"10.1109/ISSPA.2012.6310623","DOIUrl":"https://doi.org/10.1109/ISSPA.2012.6310623","url":null,"abstract":"Recent disasters have shown that there is a growing interest for remotely sensed data to support decision makers and emergency teams in the field. Fast and accurate detection of buildings and sustained damage is of great importance. Current methods rely on numerous photo-interpreters to visually analyze the data. Multiple pixel-based methods exist to classify pixels as being part of a building or not but results vary widely and precision is often poor with very high resolution images. This paper proposes an object-based solution to building detection and compares it to a traditional approach. Object-based classification clearly provides adequate results in much less time and thus is ideal for disaster response.","PeriodicalId":248763,"journal":{"name":"2012 11th International Conference on Information Science, Signal Processing and their Applications (ISSPA)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129383162","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 : 2012-07-02DOI: 10.1109/ISSPA.2012.6310584
V. Ponomaryov, Eduardo Ramos-Díaz, V. Golikov
The 2D to 3D conversion is currently a hot topic for several applications because of the 3D content lack in a new era of different hardware. The proposed algorithm in 3D reconstruction is based on the wavelets, especially on the wavelet atomic functions (WAF), which are used in the computation of the disparity maps employing multilevel decomposition, and technique of 3D visualization via color anaglyphs synthesis. Novel approach performs better in depth and spatial perception than do existing techniques, both in terms of objective SSIM criterion and based on the more subjective measure of human vision that has been confirmed in numerous simulation results obtained in synthetic images, in synthetic video sequences and in real-life video sequences.
{"title":"Automatic conversion system for 3D video generation based on wavelets","authors":"V. Ponomaryov, Eduardo Ramos-Díaz, V. Golikov","doi":"10.1109/ISSPA.2012.6310584","DOIUrl":"https://doi.org/10.1109/ISSPA.2012.6310584","url":null,"abstract":"The 2D to 3D conversion is currently a hot topic for several applications because of the 3D content lack in a new era of different hardware. The proposed algorithm in 3D reconstruction is based on the wavelets, especially on the wavelet atomic functions (WAF), which are used in the computation of the disparity maps employing multilevel decomposition, and technique of 3D visualization via color anaglyphs synthesis. Novel approach performs better in depth and spatial perception than do existing techniques, both in terms of objective SSIM criterion and based on the more subjective measure of human vision that has been confirmed in numerous simulation results obtained in synthetic images, in synthetic video sequences and in real-life video sequences.","PeriodicalId":248763,"journal":{"name":"2012 11th International Conference on Information Science, Signal Processing and their Applications (ISSPA)","volume":"295 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129848250","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 : 2012-07-02DOI: 10.1109/ISSPA.2012.6310471
S. Impedovo, Francesco Maurizio Mangini, G. Pirlo
In this paper a new clustering technique for improving off-line handwritten digit recognition is introduced. Clustering design is approached as an optimization problem in which the objective function to be minimized is the cost function associated to the classification, that is here performed by the k-nearest neighbor (k-NN) classifier based on the Sokal and Michener dissimilarity measure. For this purpose, a genetic algorithm is used to determine the best cluster centers to reduce classification time, without suffering a great loss in accuracy. In addition, an effective strategy for generating the initial-population of the genetic algorithm is also presented. The experimental tests carried out using the MNIST database show the effectiveness of this method.
{"title":"A genetic algorithm based clustering approach for improving off-line handwritten digit classification","authors":"S. Impedovo, Francesco Maurizio Mangini, G. Pirlo","doi":"10.1109/ISSPA.2012.6310471","DOIUrl":"https://doi.org/10.1109/ISSPA.2012.6310471","url":null,"abstract":"In this paper a new clustering technique for improving off-line handwritten digit recognition is introduced. Clustering design is approached as an optimization problem in which the objective function to be minimized is the cost function associated to the classification, that is here performed by the k-nearest neighbor (k-NN) classifier based on the Sokal and Michener dissimilarity measure. For this purpose, a genetic algorithm is used to determine the best cluster centers to reduce classification time, without suffering a great loss in accuracy. In addition, an effective strategy for generating the initial-population of the genetic algorithm is also presented. The experimental tests carried out using the MNIST database show the effectiveness of this method.","PeriodicalId":248763,"journal":{"name":"2012 11th International Conference on Information Science, Signal Processing and their Applications (ISSPA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130639213","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 : 2012-07-02DOI: 10.1109/ISSPA.2012.6310440
J. Orozco-Arroyave, J. Vargas-Bonilla, J. B. Alonso, M. A. Ferrer-Ballester, C. Travieso-González, P. H. Rodríguez
A novel methodology, based on the estimation of nonlinear dynamics features, is presented for automatic detection of pathologies in the phonatory system considering continuous speech records (text-dependent). The proposed automatic segmentation and characterization of the voice registers does not require the estimation of the pitch period, therefore it doesn't depend on the gender and intonation of the patients. A robust methodology for finding the features that better discriminate between healthy and pathological voices and also for analyzing the affinity among them is also presented. An average success rate of 95% ± 3.54% in the automatic detection of voice pathologies is achieved considering only six features. The results indicate that nonlinear dynamics is a good alternative for automatic detection of abnormal phonations in continuous speech.
{"title":"Voice pathology detection in continuous speech using nonlinear dynamics","authors":"J. Orozco-Arroyave, J. Vargas-Bonilla, J. B. Alonso, M. A. Ferrer-Ballester, C. Travieso-González, P. H. Rodríguez","doi":"10.1109/ISSPA.2012.6310440","DOIUrl":"https://doi.org/10.1109/ISSPA.2012.6310440","url":null,"abstract":"A novel methodology, based on the estimation of nonlinear dynamics features, is presented for automatic detection of pathologies in the phonatory system considering continuous speech records (text-dependent). The proposed automatic segmentation and characterization of the voice registers does not require the estimation of the pitch period, therefore it doesn't depend on the gender and intonation of the patients. A robust methodology for finding the features that better discriminate between healthy and pathological voices and also for analyzing the affinity among them is also presented. An average success rate of 95% ± 3.54% in the automatic detection of voice pathologies is achieved considering only six features. The results indicate that nonlinear dynamics is a good alternative for automatic detection of abnormal phonations in continuous speech.","PeriodicalId":248763,"journal":{"name":"2012 11th International Conference on Information Science, Signal Processing and their Applications (ISSPA)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124481175","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 : 2012-07-02DOI: 10.1109/ISSPA.2012.6310498
C. Srinivasa, M. Bouchard, R. Pichevar, Hossein Najaf-Zadeh
A novel framework based on graph theory for structure discovery is applied to audio to find new types of audio objects which enable the compression of an input signal. It converts the sparse time-frequency representation of an audio signal into a graph by representing each data point as a vertex and the relationship between two vertices as an edge. Each edge is labelled based on a clustering algorithm which preserves a quality guarantee on the clusters. Frequent subgraphs are then extracted from this graph, via a mining algorithm, and recorded as objects. Tests performed using a corpus of audio excerpts show that the framework discovers new types of audio objects which yield an average compression gain of 23.53% while maintaining high audio quality.
{"title":"Graph theory for the discovery of non-parametric audio objects","authors":"C. Srinivasa, M. Bouchard, R. Pichevar, Hossein Najaf-Zadeh","doi":"10.1109/ISSPA.2012.6310498","DOIUrl":"https://doi.org/10.1109/ISSPA.2012.6310498","url":null,"abstract":"A novel framework based on graph theory for structure discovery is applied to audio to find new types of audio objects which enable the compression of an input signal. It converts the sparse time-frequency representation of an audio signal into a graph by representing each data point as a vertex and the relationship between two vertices as an edge. Each edge is labelled based on a clustering algorithm which preserves a quality guarantee on the clusters. Frequent subgraphs are then extracted from this graph, via a mining algorithm, and recorded as objects. Tests performed using a corpus of audio excerpts show that the framework discovers new types of audio objects which yield an average compression gain of 23.53% while maintaining high audio quality.","PeriodicalId":248763,"journal":{"name":"2012 11th International Conference on Information Science, Signal Processing and their Applications (ISSPA)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128849754","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 : 2012-07-02DOI: 10.1109/ISSPA.2012.6310515
R. Ashrafi, J. Azaña
A novel all-optical design for implementing THz-bandwidth real-time Hilbert transformers is proposed and numerically demonstrated. We show that an all-optical Hilbert transformer can be implemented using a uniform-period long-period fiber grating (LPG) with a properly designed amplitude-only grating apodization profile incorporating a single/multiple π-phase-shift(s) along the grating length. The designed LPG for implementation of Hilbert transformer operates in the cross-coupling mode, which can be practically implemented based on either a fiber-optic approach or integrated-waveguide technology. All-optical Hilbert transformers capable of processing arbitrary optical signals with bandwidths well in the THz range can be implemented using feasible LPG designs.
{"title":"All-optical ultrafast hilbert transformations based on all-fiber long period grating designs","authors":"R. Ashrafi, J. Azaña","doi":"10.1109/ISSPA.2012.6310515","DOIUrl":"https://doi.org/10.1109/ISSPA.2012.6310515","url":null,"abstract":"A novel all-optical design for implementing THz-bandwidth real-time Hilbert transformers is proposed and numerically demonstrated. We show that an all-optical Hilbert transformer can be implemented using a uniform-period long-period fiber grating (LPG) with a properly designed amplitude-only grating apodization profile incorporating a single/multiple π-phase-shift(s) along the grating length. The designed LPG for implementation of Hilbert transformer operates in the cross-coupling mode, which can be practically implemented based on either a fiber-optic approach or integrated-waveguide technology. All-optical Hilbert transformers capable of processing arbitrary optical signals with bandwidths well in the THz range can be implemented using feasible LPG designs.","PeriodicalId":248763,"journal":{"name":"2012 11th International Conference on Information Science, Signal Processing and their Applications (ISSPA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125905556","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 : 2012-07-02DOI: 10.1109/ISSPA.2012.6310596
Messaoud Bengherabi, F. Harizi, A. Guessoum, M. Cheriet
The aim of this paper is to investigate the user-specific two-level fusion strategy in the context of multimodal biometrics. In this strategy, a client-specific score normalization procedure is applied firstly to each of the system outputs to be fused. Then, the resulting normalized outputs are fed into a common classifier. The logistic regression, non-confidence weighted sum and the likelihood ratio based on Gaussian mixture model are used as back-end classifiers. Three client-specific score normalization procedures are considered in this paper, i.e. Z-norm, F-norm and the Model-Specific Log-Likelihood Ratio MSLLR-norm. Our first findings based on 15 fusion experiments on the XM2VTS score database show that when the previous two-level fusion strategy is applied, the resulting fusion classifier outperforms the baseline classifiers significantly and a relative reduction of more than 50% in the equal error rate can be achieved. The second finding is that when using this two-level user-specific fusion strategy, the design of the final classifier is simplified and performance generalization of baseline classifiers is not straightforward. A great attention must be given to the choice of the combination normalization-back-end classifier.
{"title":"Incorporating user specific normalization in multimodal biometric fusion system","authors":"Messaoud Bengherabi, F. Harizi, A. Guessoum, M. Cheriet","doi":"10.1109/ISSPA.2012.6310596","DOIUrl":"https://doi.org/10.1109/ISSPA.2012.6310596","url":null,"abstract":"The aim of this paper is to investigate the user-specific two-level fusion strategy in the context of multimodal biometrics. In this strategy, a client-specific score normalization procedure is applied firstly to each of the system outputs to be fused. Then, the resulting normalized outputs are fed into a common classifier. The logistic regression, non-confidence weighted sum and the likelihood ratio based on Gaussian mixture model are used as back-end classifiers. Three client-specific score normalization procedures are considered in this paper, i.e. Z-norm, F-norm and the Model-Specific Log-Likelihood Ratio MSLLR-norm. Our first findings based on 15 fusion experiments on the XM2VTS score database show that when the previous two-level fusion strategy is applied, the resulting fusion classifier outperforms the baseline classifiers significantly and a relative reduction of more than 50% in the equal error rate can be achieved. The second finding is that when using this two-level user-specific fusion strategy, the design of the final classifier is simplified and performance generalization of baseline classifiers is not straightforward. A great attention must be given to the choice of the combination normalization-back-end classifier.","PeriodicalId":248763,"journal":{"name":"2012 11th International Conference on Information Science, Signal Processing and their Applications (ISSPA)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126497544","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：