|
|
|
题名
|
作者
|
年代
|
出处
|
被引量
|
| 1 | 一种稳健的基于VisemicLDA的口形动态特征及听视觉语音识别显示文摘视觉特征提取是听视觉语音识别研究的热点问题。文章引入了一种稳健的基于Visemic LDA的口形动态特征,这种特征充分考虑了发音时口形轮廓的变化及视觉Viseme划分。文章同时提出了一利利用语音识别结果进行LDA训练数据自动标注的方法。这种方法免去了繁重的人工标注工作,避免了标注错误。实验表明,将'VisemicLDA视觉特征引入到听视觉语音识别中,可以大大地提高噪声条件下语音识别系统的识别率;将这种视觉特征与多数据流HMM结合之后,在信噪比为10dB的强噪声情况下,识别率仍可以达到80%以上。 | 谢磊 付中华 蒋冬梅 赵荣椿 Werner Verhelst Hichem Sahli Jan Conlenis | 2005 | 电子与信息学报2005，27，1： | 4 |
| 2 | ON A BI-HARMONIC EQUATION INVOLVING CRITICAL EXPONENT: EXISTENCE AND MULTIPLICITY RESULTS显示文摘In this paper, we consider the problem of existence as well as multiplicity results for a bi-harmonic equation under the Navier boundary conditions: △2 u = K(x)u p , u > 0 in Ω , △u = u = 0 on Ω , where Ω is a smooth domain in R n , n 5, and p + 1 = 2 n n 4 is the critical Sobolev exponent. We obtain highlightly a new criterion of existence, which provides existence results for a dense subset of positive functions, and generalizes Bahri-Coron type criterion in dimension six. Our argument gives also estimates on the Morse index of the obtained solutions and extends some known results. Moreover, it provides, for generic K, Morse inequalities at infinity, which delivers lower bounds for the number of solutions. As further applications of this Morse theoretical approach, we prove more existence results. | Zakaria Boucheche Ridha Yacoub Hichem Chtioui | 2011 | Acta Mathematica Scientia2011，31，4： | 4 |
| 3 | 基于三音素动态贝叶斯网络模型的大词汇量连续语音识别显示文摘考虑连续语音中的协同发音现象,基于词-音素结构的DBN(WP-DBN)模型和词-音素-状态结构的DBN(WPS-DBN)模型,引入上下文相关的三音素单元,提出两个新颖的单流DBN模型:基于词-三音素结构的DBN(WT-DBN)模型和基于词-三音素-状态的DBN(WTS-DBN)模型。WTS-DBN模型是三音素模型,识别基元为三音素,以显式的方式模拟了基于三音素状态捆绑的隐马尔可夫模型(HMM)。大词汇量语音识别实验结果表明:在纯净语音环境下,WTS-DBN模型的识别率比HMM,WT-DBN,WP-DBN和WPS-DBN模型的识别率分别提高了20.53%,40.77%,42.72%和7.52%。 | 吕国云 赵荣椿 张艳宁 樊养余 Sahli Hichem | 2009 | 数据采集与处理2009，24，1： | 3 |
| 4 | 双模型语音识别中的听视觉合成和模型同步异步性实验研究显示文摘研究了双模型语音识别系统中前合成和后合成两种听觉视觉合成方法 ;同时在后合成方法中引入了考虑听觉和视觉同步异步特点的复合模型。仿真实验证明了在声学噪音环境下 ,后合成方法能够带来比较理想的识别效果 ;考虑听觉和视觉同步异步性的模型可以有效地提高识别率。 | 谢磊 蒋冬梅 Ilse Ravyse 赵荣椿 Hichem Sahli Werner Verhelst Jan Cornelis | 2004 | 西北工业大学学报2004，22，2： | 3 |
| 5 | 固体推进剂速度耦合响应函数测量实验方法研究显示文摘为获得固体推进剂速度耦合响应特性,基于T型燃烧器测量技术,建立了一种测量含铝复合推进剂速度耦合响应函数的实验方法,并在工作压强为6.5MPa和振荡频率约为175Hz条件下,开展了响应特性实验研究。实验结果表明:速度耦合与压强耦合有明显的不同,压强耦合在整个振荡区域内均表现出增益作用,而速度耦合在不同振荡区域内可表现出增益作用,也可以表现出抑制作用;在本文的工况参数条件下,速度耦合响应函数值远大于压强耦合响应函数值。在固体火箭发动机非线性燃烧不稳定中,速度耦合响应是不可忽略的重要影响因素。本文所建立的实验方法能在测量速度耦合响应函数的同时获得压强耦合响应函数,可更好地分析固体推进剂的燃烧响应特性。因此,该方法为分析固体火箭发动机非线性燃烧不稳定问题提供了重要的实验手段。 | 金秉宁 刘佩进 Hichem Rezaiguia 魏少娟 徐冠宇 | 2019 | 推进技术2019，40，1： | 3 |
| 6 | Anti-inflammatory and analgesic activities with gastroprotective effect of semi-purified fractions and isolation of pure compounds from Mediterranean gorgonian Eunicella singularis显示文摘Objective: To explore anti-inflammatory activities of organic extract and its semi-purified fractions(ethanol, acetone, methanol/dichloromethane) from the Mediterranean gorgonian Eunicella singularis. Methods: The anti-inflammatory and analgesic activities were evaluated, using the carrageenan-induced rat paw edema model and the acetic acid writhing test in mice. The gastroprotective activity was determined using HCl/Et OH induced gastric ulcers in rats. The purification and structure elucidation of compound(s) from the more effective fraction were determined by chromatographic and spectroscopic methods and in comparison with data reported in the literature. Results: The fraction F-Et OH showed an important antiinflammatory activity associated with significant analgesic and gastroprotective properties. The purification and structure elucidation of compound(s) from this fraction lead to the identification of one diterpenoid and four sterols. Conclusions: These results suggested that components from the active fraction can be used to treat various anti-inflammatory diseases. | Monia Deghrigue Carmen Festa Lotfi Ghribi Maria Valeria D’Auria Simona De Marino Hichem Ben Jannet Abderrahman Bouraoui | 2015 | Asian Pacific Journal of Tropical Medicine2015，8，8： | 2 |
| 7 | 基于发音特征DBN模型的嘴部动画合成显示文摘具有真实感的面部动画合成是虚拟现实的重要研究内容,在传统的基于多流隐马尔可夫模型(MSHMM)的可视语音合成中,由于MSHMM不能为发音器官的运动关系建模,合成的嘴部图像模糊且缺乏细节变化。提出了结合发音特征的多流动态贝叶斯网络语音模型(AF_AVDBN),定义了各节点的条件概率分布,使得发音特征(如嘴唇、舌体和声门/软腭)之间可以异步,并在此基础上推导了基于极大似然估计原理的嘴部最优参数学习算法。嘴部动画合成实验结果表明,基于AF_AVDBN的合成方法能够得到非常逼真的嘴部图像,效果远远优于基于MSHMM的可视语音合成方法,能够更好地应用于人机交互等领域。 | 刘培桢 蒋冬梅 RAVYSE Ilse SAHLI Hichem | 2010 | 科学技术与工程2010，10，14： | 2 |
| 8 | AED-Net:An Abnormal Event Detection Network显示文摘It has long been a challenging task to detect an anomaly in a crowded scene.In this paper,a selfsupervised framework called the abnormal event detection network(AED-Net),which is composed of a principal component analysis network(PCAnet)and kernel principal component analysis(kPCA),is proposed to address this problem.Using surveillance video sequences of different scenes as raw data,the PCAnet is trained to extract high-level semantics of the crowd’s situation.Next,kPCA,a one-class classifier,is trained to identify anomalies within the scene.In contrast to some prevailing deep learning methods,this framework is completely self-supervised because it utilizes only video sequences of a normal situation.Experiments in global and local abnormal event detection are carried out on Monitoring Human Activity dataset from University of Minnesota(UMN dataset)and Anomaly Detection dataset from University of California,San Diego(UCSD dataset),and competitive results that yield a better equal error rate(EER)and area under curve(AUC)than other state-of-the-art methods are observed.Furthermore,by adding a local response normalization(LRN)layer,we propose an improvement to the original AED-Net.The results demonstrate that this proposed version performs better by promoting the framework’s generalization capacity. | Tian Wang Zichen Miao Yuxin Chen Yi Zhou Guangcun Shan Hichem Snoussi | 2019 | Engineering2019，5，5： | 2 |
| 9 | Contribution of microbial phytases to the improvement of plant growth and nutrition: A review显示文摘Phytases belong to the class of phosphohydrolases that begin the step-wise hydrolysis of phosphates from phytates. Phytates are a derivative of myo-inositol, which is the primary storage form of organic phosphorus in plant cells. Phytase has been used globally to diminish phosphorus pollution and to enhance nutrition in monogastrics. In this review, the classification, sources, and diversity of microbial phytases, and their practical applications, as well as supplementation of the soil with transgenic and wild types of microbial strains, which can release phytase to enhance phosphorus availability for plant uptake and reduce the need for fertilizers, are discussed. The overexpressed microbial phytases in transgenic plants enhance the growth capacity of co-cultivated plants and can therefore be employed in agricultural and biotechnological practices, such as intercropping. The introduction of phytases into the soil for improved plant growth and enhanced crop yield can be accomplished without extra cost. A diverse group of photoautotrophic microalgae can synthesize phytase and will likely be useful in many human food and animal industries. | Bijender SINGH Ines BOUKHRIS PRAGYA Vinod KUMAR Ajar Nath YADAV Ameny FARHAT-KHEMAKHEM Anil KUMAR Davender SINGH Monia BLIBECH Hichem CHOUAYEKH Othman AALGHAMDI | 2020 | Pedosphere2020，30，3： | 2 |
| 10 | 基于Viseme的连续语音识别系统及Talking Head显示文摘为实现听觉/视觉驱动的说话人头部动画,该文给出了一个基于viseme(说话时的基本嘴形单位)的连续语音识别系统。它训练viseme隐马尔可夫模型(HMM),识别语音为viseme图像序列。建模采用triseme的概念来考虑viseme的上下文相关性,但它需要超大量的训练数据。该文根据viseme图像及其相似度权值(VSW)定义视觉问题集,用来建立triseme决策树,以实现triseme的状态捆绑及HMM参数共享。为比较系统性能,基于phoneme(听觉领域的语音基本单位)的语音识别结果也被映射为viseme序列。在评价准则上,定义viseme图像相似度加权识别精度,更全面地考虑输出和参考图像序列的差别,并用嘴形圆度和VSW曲线中的突变点来评估所得viseme序列的平滑性。结果表明,基于viseme的语音识别系统能给出更平滑和合理的嘴形图像序列。 | 蒋冬梅 谢磊 Ilse Ravyse 赵荣椿 Hichem Sahli Jan Cornelis | 2004 | 电子与信息学报2004，26，3： | 2 |
| 11 | 基于基频特征的情感语音识别研究显示文摘探索在不同的情感状态下的基音特征变化规律。通过对含有生气、高兴、悲伤情感语音信号进行分析,总结了情感语音基频的变化规律,确定了用于情感识别的12维的基频的基本特征以及扩展特征,运用混合高斯模型进行情感识别,并作了识别实验,获得了较好的结果。 | 郭鹏娟 蒋冬梅 Hichem Sahli Werner Verhelst | 2007 | 计算机应用研究2007，24，10： | 2 |
| 12 | 基于动态贝叶斯网络的语音识别及音素切分研究显示文摘研究了一种基于动态贝叶斯网络(dynamic bayesian networks,DBN)的语音识别建模方法,利用GMTK(graphical model tool kits)工具构建音素级音频流DBN语音训练和识别模型,同时与传统的基于隐马尔可夫的语音识别结果进行比较,并给出词与音素的切分结果。实验表明,在各种信噪比测试条件下,基于DBN的语音识别结果与基于HMM的语音识别结果相当,并表现出一定的抗噪性,音素的切分结果也比较准确。 | 孙阿利 蒋冬梅 吕国云 Hichem Sahli Werner Verhelst | 2007 | 计算机应用研究2007，24，10： | 2 |
| 13 | A two variable refined plate theory for the bending analysis of functionally graded plates显示文摘Bending analysis of functionally graded plates using the two variable refined plate theory is presented in this paper.The number of unknown functions involved is reduced to merely four,as against five in other shear deformation theories. The variationally consistent theory presented here has, in many respects,strong similarity to the classical plate theory. It does not require shear correction factors,and gives rise to such transverse shear stress variation that the transverse shear stresses vary parabolically across the thickness and satisfy shear stress free surface conditions.Material properties of the plate are assumed to be graded in the thickness direction with their distributions following a simple power-law in terms of the volume fractions of the constituents.Governing equations are derived from the principle of virtual displacements, and a closed-form solution is found for a simply supported rectangular plate subjected to sinusoidal loading by using the Navier method.Numerical results obtained by the present theory are compared with available solutions,from which it can be concluded that the proposed theory is accurate and simple in analyzing the static bending behavior of functionally graded plates. | Ismail Mechab Hassen Ait Atmane Abdlouahed Tounsi Hichem Abdesselem Belhadj EI Abbas Adda Bedia | 2010 | Acta Mechanica Sinica2010，26，6： | 2 |
| 14 | 基于发音特征的音视频融合语音识别模型显示文摘构建一种基于发音特征的音视频双流动态贝叶斯网络(DBN)语音识别模型(AF_AV_DBN),定义节点的条件概率关系,使发音特征状态的变化可以异步。在音视频语音数据库上的语音识别实验表明,通过调整发音特征之间的异步约束,AF_AV_DBN模型能得到比基于状态的同步和异步DBN模型以及音频单流模型更高的识别率,对噪声也具有较好的鲁棒性。 | 吴鹏 蒋冬梅 王风娜 Hichem SAHLI Werner VERHELST | 2011 | 计算机工程2011，37，22： | 2 |
| 15 | Abnormal event detection via the analysis of multi-frame optical flow information显示文摘Security surveillance of public scene is closely relevant to routine safety of individual.Under the stimulus of this concern,abnormal event detection is becoming one of the most important tasks in computer vision and video processing.In this paper,we propose a new algorithm to address the visual abnormal detection problem.Our algorithm decouples the problem into a feature descriptor extraction process,followed by an AutoEncoder based network called cascade deep AutoEncoder(CDA).The movement information is represented by a novel descriptor capturing the multi-frame optical flow information.And then,the feature descriptor of the normal samples is fed into the CDA network for training.Finally,the abnormal samples are distinguished by the reconstruction error of the CDA in the testing procedure.We validate the proposed method on several video surveillance datasets. | Tian WANG Meina QIAO Aichun ZHU Guangcun SHAN Hichem SNOUSSI | 2020 | Frontiers of Computer Science2020，14，2： | 2 |
| 16 | New antioxidant bibenzyl derivative and isoflavonoid from the Tunisian Salsola tetrandra Folsk显示文摘 | Ahlem Beyaoui Atef Chaari Hatem Ghouila M’hamed Ali Hamza Hichem Ben Jannet | 2012 | Natural Product Research2012，，3： | 1 |
| 17 | A stochastic framework for the identification of building rooftops using a single remote sensing image 显示文摘 | Katartzis Antonis Sahli Hichem | 2008 | IEEE Transactions on Geoscience and Remote Sensing2008，46，1： | 1 |
| 18 | Differential cytokine and chemokine responses of bovine mammary epithelial cells to Staphylococcus aureus and Escherichia coli 显示文摘 | Hichem L Etienne M Pascal R | 2007 | Cytokine2007，38，： | 1 |
| 19 | A Stochastic Framework for the Identification of Building Rooftops Using a Single Remote Sensing Image显示文摘 | Antonis K S Hichem | 2008 | IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING2008，46，01： | 1 |
| 20 | A Selective 3-Anchor DV-Hop Algorithm Based Onthe Nearest Anchor for Wireless Sensor Network显示文摘 | Hichem Sassi Tawfik Najeh’Noureddine Liouane | 2014 | International Scholarly and Scientific Research & Innovation2014，8，10： | 1 |
您所在的机构暂未开通此部分的功能，请联系管理员