Automatic detection of voice impairments by means of short-term cepstral parameters and neural network based detectors

It is well known that vocal and voice diseases do not necessarily cause perceptible changes in the acoustic voice signal. Acoustic analysis is a useful tool to diagnose voice diseases being a complementary technique to other methods based on direct observation of the vocal folds by laryngoscopy. Through the present paper two neural-network based classification approaches applied to the automatic detection of voice disorders will be studied. Structures studied are multilayer perceptron and learning vector quantization fed using short-term vectors calculated accordingly to the well-known Mel Frequency Coefficient cepstral parameterization. The paper shows that these architectures allow the detection of voice disorders--including glottic cancer--under highly reliable conditions. Within this context, the Learning Vector quantization methodology demonstrated to be more reliable than the multilayer perceptron architecture yielding 96% frame accuracy under similar working conditions.     

基于混合高斯模型的运动目标检测

针对现有方法在复杂的环境下不能很好地检测出运动物体的问题,提出了一种改进的基于混合高斯模型的背景消减法检测运动目标。改进了背景模型的更新算法,提高了背景更新速度。利用帧间差分法消除了运动影子和光照突变问题,并采用（r,g,l）颜色空间检测和抑制了运动阴影。实验结果表明,该算法能很好地提取出运动目标。     

Gaussian mixture PHD filter for jump Markov models based on best-fitting Gaussian approximation

A new Gaussian mixture probability hypothesis density (PHD) filter is developed for tracking multiple maneuvering targets that follow jump Markov models. This approach is based on the best-fitting Gaussian approximation which has been shown to be an accurate predictor of the interacting multiple model (IMM) performance. Compared with the existing Gaussian mixture multiple model PHD filter without interacting, simulations show that the proposed filter achieves better results with much less computational expense.     

特征降维和高斯混合模型的体育运动动作识别

为了解决当前体育运动动作识别方法的不足,以获得更优的体育运动动作识别效果,提出特征降维和高斯混合模型的体育运动动作识别方法。首先采集体育运动动作的视频图像,并提取体育运动动作的特征向量,然后采用随机投影算法对特征向量进行降维处理,最后采用高斯混合模型对降维后的训练样本进行学习,构建体育运动动作识别模型,并采用各种体育运动动作数据集对性能进行测试。结果表明,该方法获得了理想的体育运动动作识别结果,而且识别正确率高于其他体育运动动作识别方法。     

基于混合高斯模型与三帧差分的目标检测算法

针对传统目标检测方法中光照变化、复杂背景、阴影等难点,提出了一种结合三帧差分法和混合高斯背景建模的算法,既能很好地适应场景中的光照渐变和背景扰动,又能克服普通帧差法中检测目标不准确,容易产生孔洞及双影现象的问题。同时,采用了一种简易的阴影抑制算法和形态学滤波处理,有效地去除了阴影以及噪声。实验结果表明,该算法易于实现,具有较好地实时性和鲁棒性,能精确地检测出运动目标。     

On using non-linear canonical correlation analysis for voice conversion based on Gaussian mixture model

Voice conversion algorithm aims to provide high level of similarity to the target voice with an acceptable level of quality.The main object of this paper was to build a nonlinear relationship between the parameters for the acoustical features of source and target speaker using Non-Linear Canonical Correlation Analysis(NLCCA) based on jointed Gaussian mixture model.Speaker indi-viduality transformation was achieved mainly by altering vocal tract characteristics represented by Line Spectral Frequencies(LSF).T...     

基于Nb/Mc口的话音质量评估研究

随着移动通信日益普及,用户对服务品质和性能的要求也越来越高。话音业务作为用户最基本的业务,是评价用户感知质量的最重要方面。如何客观评估话音质量,高效地发现现网话音业务中存在的问题,定位问题的产生原因,都是移动通信网络维护面临的新课题。本文研究的内容,就是通过分析现网的话音媒体流,对用户通话的质量进行评估,定位话音降质的故障原因及故障点,并使媒体流的分析能够结合实际工作的需要,提供与话音质量相关的核心网、无线网支撑服务,即研究网络改造解决新方案。     

Bayesian learning of finite generalized Gaussian mixture models on images

This paper presents a fully Bayesian approach to analyze finite generalized Gaussian mixture models which incorporate several standard mixtures, widely used in signal and image processing applications, such as Laplace and Gaussian. Our work is motivated by the fact that the generalized Gaussian distribution (GGD) can be applied on a wide range of data due to its shape flexibility which justifies its usefulness to model the statistical behavior of multimedia signals [1]. We present a method to evaluate the posterior distribution and Bayes estimators using a Gibbs sampling algorithm. For the selection of number of components in the mixture, we use the integrated likelihood and Bayesian information criteria. We validate the proposed method by applying it to: synthetic data, real datasets, texture classification and retrieval, and image segmentation; while comparing it to different other approaches.     

High-quality voice conversion system based on GMM statistical parameters and RBF neural network

A voice conversion （VC） system was designed based on Gaussian mixture model （GMM） and radial basis function （RBF） neural network. As a voice conversion model, RBF network needs quantities of training data to improve its performance. For one speech, the networks trained by different segments of data have different transformation effects. Since trying segment by segment to obtain the best conversion effect is complex, a conversion method was proposed, that uses GMM for statistics before training RBF network to aim at the problem. The speech transformation and representation using adaptive interpolation of weighted spectrum （STRAIGHT） model is used for accurate extraction of vocal tract spectrum. Then GMM is used to classify the numerous spectral parameters. The obtained mean parameters were trained in RBF network. Experiment reveals that, the soft classification ability of GMM can promptly realize the reduction and classification of training data under the premise of ensuring the training effect. The selection complexity is decreased thereafter. Compared to the conventional RBF network training methods, this method can make the transformation of spectral parameters more effective and improve the quality of converted speech.     

SAR成像质量指标评估软件系统的设计

合成孔径雷达(SAR,Synthetic Aperture Radar)是一种应用广泛的微波遥感成像雷达,能够全天时、全天候对地观测成像,并且图像有着高方位分辨率。SAR图像研究是遥感方向最为重要的前沿领域之一。目前没有一套专门的软件系统工具来对SAR成像进行指标的计算与测量,都是建立在通用的辅助工具如MATLAB上,计算效率低且不方便。文中设计了一种SAR成像的质量评估系统软件,该系统中包含一套比较完备的SAR成像质量指标,然后以这套指标为主要出发点,设计开发出SAR图像质量指标评价的软件。该软件以C/C++语言为基础,以OGRE和CEGUI等开源库为框架。该系统能够实现对数据文件的读取、图像的基本操作、图像和数据的质量指标测定、指标的测定和计算等,并且有着友好的人机交互功能。     

基于OpenCV与混合高斯建模的运动目标检测

针对静态背景下的视频运动序列,在研究现有的检测算法——帧间差分法与背景差分法的基础上,进一步研究了运动目标检测中背景动态建模的方法——混合高斯建模法,在此基础上提出了基于混合高斯模型与三帧差分的运动目标检测改进算法。由于使用背景差分法检测运动目标时,运动物体和阴影都将被看作运动的目标,于是研究了基于归一化RGB色彩模型的阴影处理方法,对阴影区域进行检测与去除。然后使用计算机视觉类库OpenCV结合Visual C＋＋6．0对上述算法进行实现,取得了很好的检测效果。     

基于边缘信息和混合高斯融合的运动目标检测

针对帧间差分法在目标运动较慢时无法完整的检测轮廓,混合高斯模型易受光照影响导致目标快速运动时无法辨别轮廓等问题,提出了一种更加优化的运动目标检测算法.该算法将三帧差分法与混合高斯模型相融合,利用视频中连续的三帧图像两两作差分后作或运算、二值化、形态学处理,对中间帧的进行canny边缘检测,将两次结果再进行或运算、形态学处理后得到更加完整的轮廓.用中间帧进行高斯混合模型提取前景,二值化后和边缘信息进行与运算,经过形态学处理和孔洞填充后获得运动目标.经过实验表明,该方法能够获得更加理想的运动目标.     

Stereoscopic video quality assessment based on visual attention and just-noticeable difference models

With the consideration that incorporating visual saliency information appropriately can benefit image quality assessment metrics, this paper proposes an objective stereoscopic video quality assessment (SVQA) metric by incorporating stereoscopic visual attention (SVA) to SVQA metric. Specifically, based upon the multiple visual masking characteristics of HVS, a stereoscopic just-noticeable difference model is proposed to compute the perceptual visibility for stereoscopic video. Next, a novel SVA model is proposed to extract stereoscopic visual saliency information. Then, the quality maps are calculated by the similarity of the original and distorted stereoscopic videos’ perceptual visibility. Finally, the quality score is obtained by incorporating visual saliency information to the pooling of quality maps. To evaluate the proposed SVQA metric, a subjective experiment is conducted. The experimental result shows that the proposed SVQA metric achieves better performance in comparison with the existing SVQA metrics.     

VoLTE端到端语音质量评估及优化研究

目前,VoLTE用户投诉出现过单通、断续、音质等问题,严重影响用户的高清语音体验。VoLTE网络涉及网元众多,厂家众多,场景复杂,语音质量问题归属困难。文章针对VoLTE端到端语音质量存在的问题,基于华为端到端信令分析平台SEQ,针对多网元的信令面数据进行相关性分析及聚类分析,通过一系列数据挖掘,建立了快速有效的端到端质量评估及优化体系,确保快速有效地定位网络中存在问题。     

基于混合高斯模型和六帧差分的目标检测算法

鉴于传统混合高斯模型在光照突变、噪声干扰时鲁棒性不高,易造成检测错误等问题,提出了一种改进的视频运动目标检测算法。该算法将混合高斯模型与六帧差分算法相结合,构建了一种高效的运动目标轮廓模型,并嵌入背景替换法和动态阈值分割法提高算法的稳健性,通过连通性检测和形态学处理,得到完整的运动前景像素。不同场景的视频检测结果表明,改进算法有效克服了光照突变、噪声干扰、空洞及双影现象,与同类算法相比,具有更高的准确度和鲁棒性。     

基于混合高斯模型与核密度估计的目标检测

背景建模与目标检测是视频跟踪的重要步骤和基础,非参数核密度估计与混合高斯模型是背景建模与目标检测的经典方法。文中首先介绍了高斯模型与核密度估计的基本原理及各自的优缺点,然后提出了一种核密度-混合高斯模型级联算法,利用核密度估计快速分割前景与背景区域,再由混合高斯模型对于无法精确建模的区域进行二次判定,有效综合了二者各自的优点。仿真结果表明,该算法具有良好的实时性和鲁棒性。     

Network video quality assessment based on fuzzy inference system

The objective assessment method of network video quality is a challenge, because the video quality will be distorted by various factors, including transmission and compression. In order to improve the objective method, an objective assessment method based on fuzzy inference system of Mamdani is proposed. Firstly, six quality parametersare introduced. All the quality parameters are inputted to fuzzy logic controller system. Secondly, the outputs are used as next inputs and inferred by another fuzzy logic controller system to obtain the objective quality of network video. Lastly, the performance of proposed method is validated on four videos with different network environment. Meanwhile this method is compared with other methods. The experimental results show that the proposed method can improve the similarity between subjective and objective assessment.     

基于改进高斯混合模型的运动目标检测与跟踪

为了提高运动目标检测与跟踪的精确性与可靠性,提出一种基于改进高斯混合模型的运动目标检测与跟踪方法。首先,建立改进高斯混合背景模型,对运动目标图像进行分块处理,利用相连帧的连续性对运动目标图像的参数更新,提取完整的运动目标并进行分割;其次,将给定的当前帧像素点与目标图像进行匹配,减少高斯混合模型的分布数量和计算量,根据分块处理后的运动目标的大小、形状以及颜色信息完成运动目标全局匹配,实现运动目标的实时检测与跟踪。实验结果表明,与目前的高斯混合模型对运动目标检测与跟踪的方法相比,所提方法计算过程较为简单,具有更快的检测速度和更可靠的检测结果。     

Online nonparametric Bayesian analysis of parsimonious Gaussian mixture models and scenes clustering

The mixture model is a very powerful and flexible tool in clustering analysis. Based on the Dirichlet process and parsimonious Gaussian distribution, we propose a new nonparametric mixture framework for solving challenging clustering problems. Meanwhile, the inference of the model depends on the efficient online variational Bayesian approach, which enhances the information exchange between the whole and the part to a certain extent and applies to scalable datasets. The experiments on the scene database indicate that the novel clustering framework, when combined with a convolutional neural network for feature extraction, has meaningful advantages over other models.