首页 > 最新文献
paperpass

2009 17th European Signal Processing Conference最新文献

英文 中文
Image interpolation using optimized color transforms 图像插值使用优化的颜色变换
Pub Date : 2009-08-24 DOI: 10.5281/ZENODO.41427
E. Gershikov, M. Porat
In many demosaicing algorithms the red and the blue colors are reconstructed based on the interpolated green. Although additional information such as the statistics of the color components could be easily obtained, these algorithms are not optimized according to the available data. In this work we propose optimized color transforms for the reconstruction of the red and blue colors, based on properties of the input image. When comparing the results of the optimized algorithms to existing known techniques, we show improvement in the reconstructed images both visually and according to the SCIELAB quality measure. Based on extensive simulation results, we show and conclude that the proposed optimization techniques could be useful and instrumental in digital photography.
在许多去马赛克算法中,红色和蓝色是基于插值后的绿色重建的。虽然可以很容易地获得额外的信息,如颜色成分的统计信息,但这些算法并没有根据可用的数据进行优化。在这项工作中,我们提出了优化的颜色变换,用于重建红色和蓝色,基于输入图像的属性。当将优化算法的结果与现有已知技术进行比较时,我们显示出重建图像在视觉上和根据SCIELAB质量测量方面的改进。基于广泛的模拟结果,我们表明并得出结论,所提出的优化技术在数字摄影中是有用的和有用的。
{"title":"Image interpolation using optimized color transforms","authors":"E. Gershikov, M. Porat","doi":"10.5281/ZENODO.41427","DOIUrl":"https://doi.org/10.5281/ZENODO.41427","url":null,"abstract":"In many demosaicing algorithms the red and the blue colors are reconstructed based on the interpolated green. Although additional information such as the statistics of the color components could be easily obtained, these algorithms are not optimized according to the available data. In this work we propose optimized color transforms for the reconstruction of the red and blue colors, based on properties of the input image. When comparing the results of the optimized algorithms to existing known techniques, we show improvement in the reconstructed images both visually and according to the SCIELAB quality measure. Based on extensive simulation results, we show and conclude that the proposed optimization techniques could be useful and instrumental in digital photography.","PeriodicalId":106989,"journal":{"name":"2009 17th European Signal Processing Conference","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114138850","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}
引用次数: 1
Impostor detection using facial stereoscopic images 利用面部立体图像检测冒名顶替者
Pub Date : 2009-08-24 DOI: 10.5281/ZENODO.41625
Abdelaali Benaiss, Usman Saeed, J. Dugelay, M. Jedra
Biometric systems are prone to impostor attacks, which in case of high security applications could be critical. These attacks range from simple attacks consisting of stealing someone's password or complex as cryptographic attacks on the transmission channel. In this study we have focused on combating replay attacks, in which an impostor uses a prerecorded image of the client. The system captures stereoscopic images of the face and then decides if the images belong to a real face or a poster from depth calculated using triangulation.
生物识别系统容易受到冒名顶替者的攻击,这在高安全性的应用程序中可能是至关重要的。这些攻击的范围从窃取某人密码的简单攻击到传输通道上的复杂加密攻击。在这项研究中,我们专注于打击重放攻击,其中冒名顶替者使用客户端预先录制的图像。该系统捕获人脸的立体图像,然后根据三角测量计算的深度判断图像是属于真实的人脸还是海报。
{"title":"Impostor detection using facial stereoscopic images","authors":"Abdelaali Benaiss, Usman Saeed, J. Dugelay, M. Jedra","doi":"10.5281/ZENODO.41625","DOIUrl":"https://doi.org/10.5281/ZENODO.41625","url":null,"abstract":"Biometric systems are prone to impostor attacks, which in case of high security applications could be critical. These attacks range from simple attacks consisting of stealing someone's password or complex as cryptographic attacks on the transmission channel. In this study we have focused on combating replay attacks, in which an impostor uses a prerecorded image of the client. The system captures stereoscopic images of the face and then decides if the images belong to a real face or a poster from depth calculated using triangulation.","PeriodicalId":106989,"journal":{"name":"2009 17th European Signal Processing Conference","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114570847","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}
引用次数: 1
From BOLD-fMRI signals to the prediction of subjective pain perception through a regularization algorithm 从BOLD-fMRI信号到通过正则化算法预测主观疼痛感知
Pub Date : 2009-08-24 DOI: 10.5281/ZENODO.41444
M. Prato, S. Favilla, P. Baraldi, C. Porro, L. Zanni
Functional magnetic resonance imaging, in particular the BOLD-fMRI technique, plays a dominant role in human brain mapping studies, mostly because of its non-invasiveness and relatively high spatio-temporal resolution. The main goal of fMRI data analysis has been to reveal the distributed patterns of brain areas involved in specific functions, by applying a variety of statistical methods with model-based or data-driven approaches. In the last years, several studies have taken a different approach, where the direction of analysis is reversed in order to probe whether fMRI signals can be used to predict perceptual or cognitive states. In this study we test the feasibility of predicting the perceived pain intensity in healthy volunteers, based on fMRI signals collected during an experimental pain paradigm lasting several minutes. In particular, we introduce a methodological approach based on new regularization learning algorithms for regression problems.
功能磁共振成像,特别是BOLD-fMRI技术,在人脑成像研究中占据主导地位,主要是因为它的非侵入性和相对较高的时空分辨率。fMRI数据分析的主要目标是通过应用各种基于模型或数据驱动的统计方法来揭示涉及特定功能的大脑区域的分布模式。在过去的几年里,一些研究采取了不同的方法,其中分析的方向是相反的,以探索功能磁共振成像信号是否可以用来预测感知或认知状态。在这项研究中,我们测试了预测健康志愿者感知疼痛强度的可行性,基于在持续几分钟的实验性疼痛范式中收集的功能磁共振成像信号。特别地,我们介绍了一种基于新的正则化学习算法的方法来解决回归问题。
{"title":"From BOLD-fMRI signals to the prediction of subjective pain perception through a regularization algorithm","authors":"M. Prato, S. Favilla, P. Baraldi, C. Porro, L. Zanni","doi":"10.5281/ZENODO.41444","DOIUrl":"https://doi.org/10.5281/ZENODO.41444","url":null,"abstract":"Functional magnetic resonance imaging, in particular the BOLD-fMRI technique, plays a dominant role in human brain mapping studies, mostly because of its non-invasiveness and relatively high spatio-temporal resolution. The main goal of fMRI data analysis has been to reveal the distributed patterns of brain areas involved in specific functions, by applying a variety of statistical methods with model-based or data-driven approaches. In the last years, several studies have taken a different approach, where the direction of analysis is reversed in order to probe whether fMRI signals can be used to predict perceptual or cognitive states. In this study we test the feasibility of predicting the perceived pain intensity in healthy volunteers, based on fMRI signals collected during an experimental pain paradigm lasting several minutes. In particular, we introduce a methodological approach based on new regularization learning algorithms for regression problems.","PeriodicalId":106989,"journal":{"name":"2009 17th European Signal Processing Conference","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116945674","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}
引用次数: 0
A new channel order estimation algorithm for FIR SIMO channels 一种新的FIR SIMO信道阶数估计算法
Pub Date : 2009-08-24 DOI: 10.5281/ZENODO.41614
S. Karakütük, T. E. Tuncer
Channel order estimation problem is considered for FIR (Finite Impulse Response) modeled single-input multi-output (SIMO) communication systems. The performance of the channel estimation algorithms depends on the accuracy and robustness of the channel order estimation. A new channel order estimation algorithm with high accuracy and robustness is proposed for SIMO systems. The proposed algorithm is based on the least squares smoothing (LSS) algorithm for channel estimation. It is guaranteed to find the true channel order from finite number of samples for noise free case. Several experiments are performed and it is shown that the proposed algorithm significantly enhances the performance in channel order estimation in noisy observations. The comparisons with the alternative techniques show that the proposed method is very effective for the channel order estimation.
研究了基于有限脉冲响应模型的单输入多输出(SIMO)通信系统的信道阶数估计问题。信道估计算法的性能取决于信道阶数估计的准确性和鲁棒性。针对SIMO系统,提出了一种精度高、鲁棒性好的信道阶数估计算法。该算法基于最小二乘平滑(LSS)算法进行信道估计。在无噪声情况下,保证从有限个样本中找到真实的信道阶数。实验结果表明,该算法显著提高了在噪声观测条件下信道阶数估计的性能。与其他方法的比较表明,该方法对信道阶数估计是非常有效的。
{"title":"A new channel order estimation algorithm for FIR SIMO channels","authors":"S. Karakütük, T. E. Tuncer","doi":"10.5281/ZENODO.41614","DOIUrl":"https://doi.org/10.5281/ZENODO.41614","url":null,"abstract":"Channel order estimation problem is considered for FIR (Finite Impulse Response) modeled single-input multi-output (SIMO) communication systems. The performance of the channel estimation algorithms depends on the accuracy and robustness of the channel order estimation. A new channel order estimation algorithm with high accuracy and robustness is proposed for SIMO systems. The proposed algorithm is based on the least squares smoothing (LSS) algorithm for channel estimation. It is guaranteed to find the true channel order from finite number of samples for noise free case. Several experiments are performed and it is shown that the proposed algorithm significantly enhances the performance in channel order estimation in noisy observations. The comparisons with the alternative techniques show that the proposed method is very effective for the channel order estimation.","PeriodicalId":106989,"journal":{"name":"2009 17th European Signal Processing Conference","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114802937","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}
引用次数: 1
Investigation of parameter effects and truncated multistatic data matrix in decomposition of time reversal operator method 时间反转算子分解中参数效应及截短多静态数据矩阵的研究
Pub Date : 2009-08-24 DOI: 10.5281/ZENODO.41359
Dinh-Quy Nguyen, XueXin Yap, W. Gan, Y. Chong
Acoustic waves are commonly used in the detection, localization and focusing fields. DORT, a French acronym for decomposition of time reversal operator, is a novel method in which active detection and focusing of acoustic waves using arrays of transceivers is performed. The DORT method extracts relevant information from collected data and forms a multistatic data matrix (MDM) that enables the detection and focusing of signals on scattering objects. In this paper, the parameters of the DORT method are studied and their effects on the focusing performance are investigated. In addition, a truncated MDM method, which achieves both time and storage saving over the full MDM method, is proposed.
声波在探测、定位和聚焦领域中有着广泛的应用。DORT是分解时间反转算子的法语首字母缩略词,是一种利用收发器阵列进行声波主动探测和聚焦的新方法。DORT方法从采集到的数据中提取相关信息,形成一个多静态数据矩阵(MDM),实现对散射物体的信号检测和聚焦。本文研究了DORT方法的参数及其对聚焦性能的影响。此外,还提出了一种截断的MDM方法,该方法比完整的MDM方法节省时间和存储。
{"title":"Investigation of parameter effects and truncated multistatic data matrix in decomposition of time reversal operator method","authors":"Dinh-Quy Nguyen, XueXin Yap, W. Gan, Y. Chong","doi":"10.5281/ZENODO.41359","DOIUrl":"https://doi.org/10.5281/ZENODO.41359","url":null,"abstract":"Acoustic waves are commonly used in the detection, localization and focusing fields. DORT, a French acronym for decomposition of time reversal operator, is a novel method in which active detection and focusing of acoustic waves using arrays of transceivers is performed. The DORT method extracts relevant information from collected data and forms a multistatic data matrix (MDM) that enables the detection and focusing of signals on scattering objects. In this paper, the parameters of the DORT method are studied and their effects on the focusing performance are investigated. In addition, a truncated MDM method, which achieves both time and storage saving over the full MDM method, is proposed.","PeriodicalId":106989,"journal":{"name":"2009 17th European Signal Processing Conference","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115645912","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}
引用次数: 1
Multivariable statistical classification of 3D bone microarchitecture using morphological and mechanical features 基于形态学和力学特征的三维骨微结构多变量统计分类
Pub Date : 2009-08-24 DOI: 10.5281/ZENODO.41542
R. Jennane, G. Aufort, H. Lheritier, C. Benhamou
Evaluating trabecular bone microarchitecture is not possible in the clinical routine nowadays. Based on simplified skeleton and Finite Element models, we have developed different image processing and simulation techniques to investigate bone microarchitecture and its mechanical stiffness. This work defines improved models for 3D bone architecture that generates multiple features. These parameters can be combined in a statistical discriminant analysis in order to study bone diseases such as osteoporosis. A clinical study is led on 2 populations of arthritic and osteoporotic bone samples. The results show the ability of our improved techniques to discriminate the 2 populations.
目前在临床常规中还不可能对骨小梁的微结构进行评估。基于简化的骨骼模型和有限元模型,我们开发了不同的图像处理和仿真技术来研究骨骼微结构及其机械刚度。这项工作定义了生成多个特征的3D骨结构的改进模型。这些参数可以结合在统计判别分析中,以便研究骨质疏松症等骨病。对2个人群的关节炎和骨质疏松症骨样本进行了临床研究。结果表明,改进的技术能够区分这两个种群。
{"title":"Multivariable statistical classification of 3D bone microarchitecture using morphological and mechanical features","authors":"R. Jennane, G. Aufort, H. Lheritier, C. Benhamou","doi":"10.5281/ZENODO.41542","DOIUrl":"https://doi.org/10.5281/ZENODO.41542","url":null,"abstract":"Evaluating trabecular bone microarchitecture is not possible in the clinical routine nowadays. Based on simplified skeleton and Finite Element models, we have developed different image processing and simulation techniques to investigate bone microarchitecture and its mechanical stiffness. This work defines improved models for 3D bone architecture that generates multiple features. These parameters can be combined in a statistical discriminant analysis in order to study bone diseases such as osteoporosis. A clinical study is led on 2 populations of arthritic and osteoporotic bone samples. The results show the ability of our improved techniques to discriminate the 2 populations.","PeriodicalId":106989,"journal":{"name":"2009 17th European Signal Processing Conference","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116898102","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}
引用次数: 2
Making better biometric decisions with quality and cohort information: A case study in fingerprint verification 利用质量和群体信息做出更好的生物识别决策:指纹验证的案例研究
Pub Date : 2009-08-24 DOI: 10.5281/ZENODO.41648
N. Poh, A. Merati, J. Kittler
Automatically recognizing humans using their biometric traits such as face and fingerprint will have very important implications in our daily lives. This problem is challenging because biometric traits can be affected by the acquisition process which is sensitive to the environmental conditions (e.g., lighting) and the user interaction. It has been shown that post-processing the classifier output, so called score normalization, is an important mechanism to counteract the above problem. In the literature, two dominant research directions have been explored: cohort normalization and quality-based normalization. The first approach relies on a set of competing cohort models, essentially making use of the resultant cohort scores. A well-established example is the T-norm. In the second approach, the normalization is based on deriving the quality information from the raw biometric signal. We propose to combine both the cohort score- and signal-derived information via logistic regression. Based on 12 independent fingerprint experiments, our proposal is found to be significantly better than the T-norm and two recently proposed cohort-based normalization methods.
利用人脸、指纹等生物特征自动识别人类,将对我们的日常生活产生非常重要的影响。这个问题是具有挑战性的,因为生物特征可以受到采集过程的影响,这是敏感的环境条件(例如,照明)和用户交互。研究表明,对分类器输出进行后处理,即所谓的分数归一化,是解决上述问题的重要机制。在文献中,主要探索了两个研究方向:队列归一化和基于质量的归一化。第一种方法依赖于一组相互竞争的队列模型,本质上是利用所得的队列分数。一个公认的例子是t规范。在第二种方法中,归一化是基于从原始生物特征信号中提取质量信息。我们建议通过逻辑回归将队列评分和信号衍生信息结合起来。基于12个独立的指纹实验,我们的方法明显优于T-norm和最近提出的两种基于队列的归一化方法。
{"title":"Making better biometric decisions with quality and cohort information: A case study in fingerprint verification","authors":"N. Poh, A. Merati, J. Kittler","doi":"10.5281/ZENODO.41648","DOIUrl":"https://doi.org/10.5281/ZENODO.41648","url":null,"abstract":"Automatically recognizing humans using their biometric traits such as face and fingerprint will have very important implications in our daily lives. This problem is challenging because biometric traits can be affected by the acquisition process which is sensitive to the environmental conditions (e.g., lighting) and the user interaction. It has been shown that post-processing the classifier output, so called score normalization, is an important mechanism to counteract the above problem. In the literature, two dominant research directions have been explored: cohort normalization and quality-based normalization. The first approach relies on a set of competing cohort models, essentially making use of the resultant cohort scores. A well-established example is the T-norm. In the second approach, the normalization is based on deriving the quality information from the raw biometric signal. We propose to combine both the cohort score- and signal-derived information via logistic regression. Based on 12 independent fingerprint experiments, our proposal is found to be significantly better than the T-norm and two recently proposed cohort-based normalization methods.","PeriodicalId":106989,"journal":{"name":"2009 17th European Signal Processing Conference","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114705704","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}
引用次数: 20
Alternative approach for computing the activation factor of the PNLMS algorithm 计算PNLMS算法激活因子的替代方法
Pub Date : 2009-08-24 DOI: 10.5281/ZENODO.41586
Francisco das Chagas de Souza, O. J. Tobias, R. Seara, D. Morgan
This paper presents a proportionate normalized least-mean-square (PNLMS) algorithm using an individual activation factor for each adaptive filter coefficient. Such strategy is used instead of a global activation factor as in the standard PNLMS algorithm. The proposed individual activation factors, determined in terms of the corresponding adaptive filter coefficients, lead to a better distribution of the adaptation energy over the filter coefficients than the standard PNLMS does. Thereby, for impulse responses exhibiting high sparseness, the proposed algorithm achieves faster convergence, outperforming both the PNLMS and improved PNLMS (IPNLMS) algorithms.
本文提出了一种比例归一化最小均方(PNLMS)算法,该算法对每个自适应滤波系数使用单独的激活因子。该策略代替了标准PNLMS算法中的全局激活因子。与标准PNLMS相比,所提出的单个激活因子根据相应的自适应滤波器系数确定,导致自适应能量在滤波器系数上的更好分布。因此,对于具有高稀疏性的脉冲响应,该算法实现了更快的收敛,优于PNLMS和改进的PNLMS (IPNLMS)算法。
{"title":"Alternative approach for computing the activation factor of the PNLMS algorithm","authors":"Francisco das Chagas de Souza, O. J. Tobias, R. Seara, D. Morgan","doi":"10.5281/ZENODO.41586","DOIUrl":"https://doi.org/10.5281/ZENODO.41586","url":null,"abstract":"This paper presents a proportionate normalized least-mean-square (PNLMS) algorithm using an individual activation factor for each adaptive filter coefficient. Such strategy is used instead of a global activation factor as in the standard PNLMS algorithm. The proposed individual activation factors, determined in terms of the corresponding adaptive filter coefficients, lead to a better distribution of the adaptation energy over the filter coefficients than the standard PNLMS does. Thereby, for impulse responses exhibiting high sparseness, the proposed algorithm achieves faster convergence, outperforming both the PNLMS and improved PNLMS (IPNLMS) algorithms.","PeriodicalId":106989,"journal":{"name":"2009 17th European Signal Processing Conference","volume":"103 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115456971","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}
引用次数: 20
A construction of compact MFCC-type features using short-time statistics for applications in audio segmentation 基于短时统计的紧凑mfcc型特征的构建,用于音频分割
Pub Date : 2009-08-24 DOI: 10.5281/ZENODO.41364
Dirk von Zeddelmann, F. Kurth
In this paper, we propose a new class of audio feature that is derived from the well-known mel frequency cepstral coefficients (MFCCs) which are widely used in speech processing. More precisely, we calculate suitable short-time statistics during the MFCC computation to obtain smoothed features with a temporal resolution that may be adjusted depending on the application. The approach was motivated by the task of audio segmentation where the classical MFCCs, having a fine temporal resolution, may result in a high amount of fluctuations and, consequently, an unstable segmentation. As a main contribution, our proposed MFCC-ENS (MFCC-Energy Normalized Statistics) features may be adapted to have a lower, and more suitable, temporal resolution while summarizing the essential information contained in the MFCCs. Our experiments on the segmentation of radio programmes demonstrate the benefits of the newly proposed features.
在本文中,我们提出了一类新的音频特征,该特征来源于广泛应用于语音处理的mel频率倒谱系数(MFCCs)。更准确地说,我们在MFCC计算过程中计算合适的短时统计数据,以获得具有时间分辨率的平滑特征,该分辨率可以根据应用进行调整。该方法的动机是音频分割的任务,其中经典的mfc具有良好的时间分辨率,可能导致大量的波动,从而导致不稳定的分割。作为主要贡献,我们提出的MFCC-ENS (MFCC-Energy Normalized Statistics)特征可以被调整为具有更低、更合适的时间分辨率,同时总结mfcc中包含的基本信息。我们对广播节目分割的实验证明了新提出的特征的好处。
{"title":"A construction of compact MFCC-type features using short-time statistics for applications in audio segmentation","authors":"Dirk von Zeddelmann, F. Kurth","doi":"10.5281/ZENODO.41364","DOIUrl":"https://doi.org/10.5281/ZENODO.41364","url":null,"abstract":"In this paper, we propose a new class of audio feature that is derived from the well-known mel frequency cepstral coefficients (MFCCs) which are widely used in speech processing. More precisely, we calculate suitable short-time statistics during the MFCC computation to obtain smoothed features with a temporal resolution that may be adjusted depending on the application. The approach was motivated by the task of audio segmentation where the classical MFCCs, having a fine temporal resolution, may result in a high amount of fluctuations and, consequently, an unstable segmentation. As a main contribution, our proposed MFCC-ENS (MFCC-Energy Normalized Statistics) features may be adapted to have a lower, and more suitable, temporal resolution while summarizing the essential information contained in the MFCCs. Our experiments on the segmentation of radio programmes demonstrate the benefits of the newly proposed features.","PeriodicalId":106989,"journal":{"name":"2009 17th European Signal Processing Conference","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114849290","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}
引用次数: 1
New accept/reject methods for independent sampling from posterior probability distributions 从后验概率分布中独立抽样的新接受/拒绝方法
Pub Date : 2009-08-24 DOI: 10.5281/ZENODO.41808
Luca Martino, J. Míguez
Rejection sampling (RS) is a well-known method to generate (pseudo-)random samples from arbitrary probability distributions that enjoys important applications, either by itself or as a tool in more sophisticated Monte Carlo techniques. Unfortunately, the use of RS techniques demands the calculation of tight upper bounds for the ratio of the target probability density function (pdf) over the proposal density from which candidate samples are drawn. Except for the class of log-concave target pdf's, for which an efficient algorithm exists, there are no general methods to analytically determine this bound, which has to be derived from scratch for each specific case. In this paper, we tackle the general problem of applying RS to draw from an arbitrary posterior pdf using the prior density as a proposal function. This is a scenario that appears frequently in Bayesian signal processing methods. We derive a general geometric procedure for the calculation of upper bounds that can be used with a broad class of target pdf's, including scenarios with correlated observations, multimodal and/or mixture measurement noises. We provide some simple numerical examples to illustrate the application of the proposed techniques.
拒绝抽样(RS)是一种众所周知的从任意概率分布生成(伪)随机样本的方法,无论是单独使用还是作为更复杂的蒙特卡罗技术的工具,都具有重要的应用。不幸的是,RS技术的使用需要计算目标概率密度函数(pdf)与抽取候选样本的建议密度之比的严格上界。除了存在一种有效算法的log-凹目标函数外,没有一般的方法来解析确定该边界,必须针对每种特定情况从头开始推导。在本文中,我们使用先验密度作为建议函数来解决应用RS从任意后验pdf中提取的一般问题。这是在贝叶斯信号处理方法中经常出现的场景。我们推导了一个计算上界的一般几何程序,该程序可用于广泛的目标pdf,包括具有相关观测,多模态和/或混合测量噪声的场景。我们提供了一些简单的数值例子来说明所提出的技术的应用。
{"title":"New accept/reject methods for independent sampling from posterior probability distributions","authors":"Luca Martino, J. Míguez","doi":"10.5281/ZENODO.41808","DOIUrl":"https://doi.org/10.5281/ZENODO.41808","url":null,"abstract":"Rejection sampling (RS) is a well-known method to generate (pseudo-)random samples from arbitrary probability distributions that enjoys important applications, either by itself or as a tool in more sophisticated Monte Carlo techniques. Unfortunately, the use of RS techniques demands the calculation of tight upper bounds for the ratio of the target probability density function (pdf) over the proposal density from which candidate samples are drawn. Except for the class of log-concave target pdf's, for which an efficient algorithm exists, there are no general methods to analytically determine this bound, which has to be derived from scratch for each specific case. In this paper, we tackle the general problem of applying RS to draw from an arbitrary posterior pdf using the prior density as a proposal function. This is a scenario that appears frequently in Bayesian signal processing methods. We derive a general geometric procedure for the calculation of upper bounds that can be used with a broad class of target pdf's, including scenarios with correlated observations, multimodal and/or mixture measurement noises. We provide some simple numerical examples to illustrate the application of the proposed techniques.","PeriodicalId":106989,"journal":{"name":"2009 17th European Signal Processing Conference","volume":"271 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121358155","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}
引用次数: 0
期刊
2009 17th European Signal Processing Conference
全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.
﹀
mjeditor
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
Book学术官方微信
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1