针对多种处理痕迹的数字语音取证算法

现有的数字语音取证研究主要集中于对单一的某种操作进行检测，无法对不相关的操作进行判断。针对该问题，提出了一种能够同时检测经过变调、低通滤波、高通滤波和加噪这四种操作的数字语音取证方法。首先，计算语音的归一化梅尔频率倒谱系数（MFCC）统计矩特征；然后通过多个二分类器对特征进行训练，并组合投票得到多分类器；最后使用该多分类器对待测语音进行分类。在TIMIT以及UME语音库上的实验结果表明，归一化MFCC统计矩特征在库内实验中均达到了97%以上的检测率，且在对MP3压缩鲁棒性测试的实验中，检测率仍能保持在96%以上。     

Detection clustering analysis algorithm and system parameters study of the near-point multi-class foreign fiber

In this paper, we investigate a detection algorithm for foreign fiber during the processing of cotton textile. By collecting a large number of samples, we determine the color model and establish its characteristic parameters of a foreign fiber. We found foreign fibers of multiple types (classes) and proposed a classification–recognition algorithm based on clustering analysis. The maximum error of the studied recognition algorithm is 0.012, which meets the requirement to recognize foreign fibers. Through many experiments, the optimal parameters for the foreign fiber detection system were determined, and the fiber recognition rates for different types were obtained. The lowest recognition rate is 85%. This is sufficiently high to reject foreign fibers and reach the standards of the textile industry. Experimental results show that foreign fiber clustering analysis algorithm is feasible, and it not only improves the quality of foreign fiber detection significantly, but also has high theoretical value and practical value.     

Visual abstraction of dynamic network via improved multi-class blue noise sampling

Massive sequence view (MSV) is a classic timeline-based dynamic network visualization approach. However, it is vulnerable to visual clutter caused by overlapping edges, thereby leading to unexpected misunderstanding of time-varying trends of network communications. This study presents a new edge sampling algorithm called edge-based multi-class blue noise (E-MCBN) to reduce visual clutter in MSV. Our main idea is inspired by the multi-class blue noise (MCBN) sampling algorithm, commonly used in multi-class scatterplot decluttering. First, we take a node pair as an edge class, which can be regarded as an analogy to classes in multi-class scatterplots. Second, we propose two indicators, namely, class overlap and inter-class conflict degrees, to measure the overlapping degree and mutual exclusion, respectively, between edge classes. These indicators help construct the foundation of migrating the MCBN sampling from multi-class scatterplots to dynamic network samplings. Finally, we propose three strategies to accelerate MCBN sampling and a partitioning strategy to preserve local high-density edges in the MSV. The result shows that our approach can effectively reduce visual clutters and improve the readability of MSV. Moreover, our approach can also overcome the disadvantages of the MCBN sampling (i.e., long-running and failure to preserve local high-density communication areas in MSV). This study is the first that introduces MCBN sampling into a dynamic network sampling.     

基于容错度自适应支持向量机的液压泵故障诊断

针对已有在线故障诊断方法在数据量大、噪声强条件下分类速度较低、分类精度不够高等问题,结合液压泵故障类别数目大、工作环境恶劣的特点,提出了一种适用于混凝土泵车液压泵在线诊断的状态识别算法——容错度自适应支持向量机。该方法主要从四个方面对分类速度做了改进：①引入容错度因子进行模型训练;②优先选择能将某一类故障样本单独分离出来的二分类器;③在满足②的基础上选择平均支持向量机少的分类器;④引入增量学习算法对参数进行自适应调整,提高多故障诊断中对新故障类别和新故障数据的适应性,保证系统的分类精度。通过对混凝土泵车的液压泵故障诊断,证明了该方法在明显提高分类速度的同时保证了较高的分类精度。     

阻抗函数不对称下多用户类型和多模式时变拥挤收费模型

研究了公交车和私家车相互作用下的排队网络在高峰时段随时间变化的最优拥挤收费问题.将时空拓展网络(STEN)与传统的网络平衡模型技术相结合,建立了多用户类型、多模式和多准则交通网络平衡模型.不同类型的用户具有不同的时间价值(VOTs),同一类用户根据出行时间和成本的不同权重确定出行负效用或一般出行成本.此外,将对称成本函数模型进行拓展,考虑公交车和私家车间的相互作用.发现当系统目标函数采用费用作为计量单位时,存在统一的路段收费模式,使多用户类型、多模式和多准则用户平衡流转变为系统最优流模式.此外,发现阻抗函数对称的边际成本收费模型不能反应不同路段上交通流的相互作用,得到的收费值偏小.     

Multi-Class Classification Methods of Cost-Conscious LS-SVM for Fault Diagnosis of Blast Furnace

 Aiming at the limitations of rapid fault diagnosis of blast furnace, a novel strategy based on cost-conscious least squares support vector machine (LS-SVM) is proposed to solve this problem. Firstly, modified discrete particle swarm optimization is applied to optimize the feature selection and the LS-SVM parameters. Secondly, cost-conscious formula is presented for fitness function and it contains in detail training time, recognition accuracy and the feature selection. The CLS-SVM algorithm is presented to increase the performance of the LS-SVM classifier. The new method can select the best fault features in much shorter time and have fewer support vectors and better generalization performance in the application of fault diagnosis of the blast furnace. Thirdly, a gradual change binary tree is established for blast furnace faults diagnosis. It is a multi-class classification method based on center-of-gravity formula distance of cluster. A gradual change classification percentage is used to select sample randomly. The proposed new method raises the speed of diagnosis, optimizes the classification accuracy and has good generalization ability for fault diagnosis of the application of blast furnace.     

基于粒子群算法的决策树SVM多分类方法研究

针对SVM多分类问题提出了一种基于粒子群算法的最优决策树SVM生成算法,以解决传统支持向量机多分类方法存在的不可分区域和误差积累现象。该方法利用自变异的PSO聚类算法在每一决策节点自动寻找最优或近优分类决策,将数据集划分为两类,直至叶子节点为止,最终根据最优决策树构建SVM多分类结构,训练各个节点SVM分类器。将该算法应用于图像人群密度分类问题,仿真实验表明,分类精度和分类时间得到明显改善,是一种有效地的多分类算法。     

基于二叉树的LS-WSVM早期火灾多类分类研究

针对早期火灾信息特点,提出了一种基于二叉树的最小二乘小波支持向量机（Least squareswavelet support vector machine,LS-WSVM）多类分类方法.该方法首先把主成份分析用于早期火灾信息的特征提取.然后,把二叉树结构和LS-WSVM相结合,提出了基于二叉树的LS-WSVM多类分类模型,不仅避免了盲目分类和不可分情况,而且提高了分类速度和泛化能力.最后,用该模型对特征信息进行处理,从而实现了对早期火灾的多类识别.早期火灾分类实验结果表明,该方法比采用径向基核函数的最小二乘支持向量机多类分类方法具有更好的识别效果和更快的分类速度.     

次序二叉树支持向量机多类故障诊断算法研究

构建二叉树支持向量机时，如果随机地将分类器分布在二叉树的各个结点上，是不能充分发挥其性能的。考虑到样本的分布情况对分类器推广能力具有较大影响，提出一种次序二叉树支持向量机多类算法，采用样本分布半径和样本分布距离估算各个类别的样本在高维特征空间中的分布情况，把分布半径较大的类别或者分布距离较大的类别较早地分出来，并且在特征空间中给其划分较大的分类区域。转子多故障诊断实验表明，该算法的诊断速度快，故障识别率高，推广能力强，更加适合于实际故障诊断应用。     

用电客户信用评价方法研究

针对用电客户信用评价的现状，借鉴商业银行信用等级划分标准，对用电客户信用进行等级界定。采用多分类支持向量机方法以及仿真数据进行的实证分析，证明这种用电客户信用评价方法具有较高的分类正确率。