基于小波-回归支持向量机算法的风电场风速预测

将小波多分辨率分析特点和回归支持向量机算法良好的泛化性能相结合,建立小波-回归支持向量机风速预测模型。先将原始风速序列经小波分解成轮廓分量和细节分量,再对各分量分别应用支持向量机模型进行预测,最后将各分量的预测结果经小波重构得到原始风速序列的预测值。仿真表明该方法能够改善预测滞后现象以及减小突变点误差,从而提高模型的泛化性能和预测精度。     

PCA故障检测方法及应用

基于主元分析(PCA)的传感器故障检测方法中T2和SPE统计量是两个重要指标。首先介绍T2统计量超限而SPE没超限故障检测的方法。利用主元相关变量残差统计量代替平方预测误差SPE统计量,并采用累积方差贡献率确定PCA模型的主元数。该方法避免了SPE统计量的保守性。最后将该方法应用于电厂某机组工作过程检测中,通过仿真验证该方法的有效性。     

基于链路分析的作者合著关系预测研究——以图情领域为例

作者合著关系的预测对于提高科研合作效率和有效的科研管理具有重要的意义。本文以中国知网中图书情报领域核心期刊作为信息来源,获取15年（2001-2015）的文献信息。通过计算指标方差和指标性质确定对合著关系预测的指标体系,同时对比基于单指标的无监督方法和基于分类算法的监督式机器学习方法（逻辑回归、支持向量机和随机森林）的预测效果,本文最终确定基于随机森林和指标体系所构造的合著关系预测模型。通过实例应用证明该模型具有较好的准确性和稳定性。     

基于小波域的视频关键帧选择算法

视频关键帧的选择算法基于小波变换（discrete wavelet transform,DWT）特征、多分辨率处理及统计技术。算法在不同分辨率下结合视觉掩蔽特性、视频帧相关性分析小波系数并统计,根据统计结果选择合适的阈值,最后通过比较选择出视频关键帧。结果表明,该算法能选择视频关健帧,选择效果优于以像素估计、块估计等与图像内容无关的算法,且简单易行。     

面向空间通信视频编码算法的业务流模型

利用Hadamard变换的去相关性能和3D小波视频编码的帧结构特性,提出了一种新的面向空间通信视频编码算法的业务流模型。实验结果表明,新模型产生的仿真数据能够匹配视频源数据的概率密度函数,能够模拟空间通信视频源数据的短时相关性和长时相关性,在漏桶评估实验中,模型仿真数据与视频源数据的排队性能匹配效果好,可以作为空间通信系统研究的视频源模型。     

基于独立分量分析与小波神经网络的时间序列预测

对时间序列的预测是一项重要的数据挖掘技术。本文将独立分量分析方法和小波神经网络相结合,建立一种ICA—WNN预测模型,并应用于风力发电功率时间序列预测。仿真结果表明所建模型具有较好的泛化性能,得到了较高的预测精度。     

基于隐马尔可夫模型的人像鉴别算法

提出了基于小波变换和隐马尔可夫模型的人像鉴别算法. 该算法首先对图像进行3级小波分解，然后把3个不同分辨率的低频子图像由小到大排列成树状结构，形成低频小波树. 接着利用独立元分析对每个小波树枝进行去相关、降维，形成特征小波树枝，并把它作为观测向量对隐马尔可夫模型进行训练，把优化的模型参数用于人脸识别. 分析了观测向量维数与识别率的关系，以及状态个数和高斯概率混合成分的个数对识别率的影响，定性描述了隐马尔可夫模型的本质. 在ORL人脸数据库上，同其他四种相关方法进行了比较，实验结果表明，该方法识别率较高，工程上易于应用.     

LWT–DSR based new robust framework for watermark extraction under intentional attack conditions

This paper proposes a new watermarking approach using dynamic stochastic resonance (DSR) based tuning operation to extract the watermark logo from the watermarked image that has undergone different intentional and signal processing attacks. This method is intended to provide remedies from the shortcomings of the technique proposed by Lin et al. (2008), and invalidates the effect of intentional attacks recently designed by Meerwald et al. (2009). The algorithm incorporates three level image decomposition using lifting wavelet transform (LWT) and low-pass subband is utilized for data hiding purpose. Watermark bits are embedded into the blocks of non-overlapped wavelet coefficients of the cover image by quantizing the two maximum coefficients of the corresponding block. In watermark extraction process, the DSR is applied by performing the tuning operation on coefficient blocks of attacked watermarked image. It is a parameter dependent approach that enhances the performance of watermark extraction, where the parameters of DSR inherently depend on the image properties such as standard deviation or variance. As far as security is concerned, the randomization of wavelet coefficients, blocks, and watermark bits helps the framework to be more secure. The proposed technique is also examined against multiple watermarking attack and successfully proves its authenticity and ownership. Comparison of the proposed technique with recent techniques shows remarkable improvement in terms of robustness and security against various intentional, signal processing, and geometrical attacks.     

小波变换在图像融合中应用探讨

作为像素级融合的方法之一,小波变换具有多分辨分析和多尺度分析的特点,目前被广泛应用于红外和可见光图像融合中。本文通过深入研究现有的图像融合算法,研究小波基的提升方案,并对其提升处理后的小波性能进行比较。     

An adaptive high-voltage direct current detection algorithm using cognitive wavelet transform

This paper proposes an algorithm that uses wavelet level adaptive decision-making for detecting high-voltage direct current (HVDC) discharge in wavelet transform cognitively. The identification and detection of HVDC discharge is an essential area of investigation, which contributes to ensuring pipeline safety and the optimal operation of an electrical power system. The proposed algorithm overcomes the wavelet packet transform’s disadvantage of needing to determine the level in advance. The decomposition level of wavelet packet transform is controlled by calculating relative wavelet energy change to decide its wavelet level. Our proposal extracts richer features of HVDC discharge by comparing other feature extraction algorithms. To select the best-suited mother wavelet function, we also design a selection method based on quantitative and qualitative approaches. An additional objective of this study is to detect the phenomenon of HVDC discharge using CP time-series data to assess the corrosion of energy pipelines. Moreover, a third primary discovery is that a wavelet-based application framework is designed to detect the HVDC discharge and further protect the energy pipeline. These discoveries can be valuably applied to the protection of power systems. They also provide brighter perspectives on future opportunities to expand on studies-to-date on the detection and classification of time-series data.     

基于小波变换系数门限值的去噪方法研究

本文系统分析了基于小波变换的门限值去噪方法,研究了小波变换门限值选择的准则及其算法.通过计算机仿真,比较了各种准则的性能,验证了基于小波变换门限值去噪方法的滤波效果.仿真结果表明,基于小波变换门限值的去噪方法可以有效地去除宽带噪声.     

利用自适应小波变换检测心电QRS波算法的实现

为了解决心电图QRS波检测的问题,研究了一种基于自适应小波变换来检测QRS波的算法。该方法用心电信号的小波变换作为自适应白化滤波器的输入,然后对白化滤波后的输出进行匹配滤波和阈值检测来识别出QRS波。采用该方法,有利于减小检测的误检率以及提高运行速度。     

基于改进的分类矢量量化的图像压缩算法研究

主要研究了图像压缩精确度问题。针对传统的图像压缩算法精度低问题,本文提出了利用小波变换后的不同的小波系数的相关性,构造图像的特征的跨频带矢量,并对其进行分类,最后通过实验验证分析算法的有效性和具有一定的实际应用价值。     

基于小波变换的全景图像超分辨率增强算法

在研究和分析小波变换理论的基础上提出了一种基于小波变换的全景图像超分辨率增强算法,该算法充分利用了小波多分辨率分解思想,体现图像降低的自然过程;通过高分辨率小波系数,经插值逆变换可得到重构的高分辨率图像。实验结果证明该算法克服了传统的插值算法致使图像高分辨部分损失、细节被模糊的缺点,是超分辨率图像处理的一种行之有效的途径。     

Exploiting the categorical reliability difference for binary classification

In binary pattern classification, the reliabilities of statistics obtained from the samples of the two categories are generally different. When the statistics are used for modeling a classifier, such reliability difference could impact the generalization performance. We formulate a disparity index to show the statistical disparity based on the generalized eigenvalue decomposition of the categorical moment matrices. It is shown that this disparity index can effectively indicate the reliability difference between the two categories. The obtained reliability difference is subsequently utilized to adjust the regularization term of a classifier for effective learning generalization. Our experiments based on 10 real-world benchmark data sets validate the effectiveness of the proposed method.     

基于主成分分析的北京市国际科技合作基地国际科技合作能力评价模型研究

结合2014年度至2017年度185家北京市国际科技合作基地统计数据,基于主成分分析法运用Matlab对北京市国际科技合作基地国际科技合作能力进行了分析研究,构建了国际科技合作能力评价模型,并针对评价模型结果提出了相关建议,进一步推动北京市国际科技合作基地发展,提升北京国际科技合作水平。     

一种新的基于复小波变换的数字图像水印算法

本文提出了一种基于复小波变换的新的具有鲁棒性的数字图像水印方法。为了改进算法的鲁棒性，本文采用了纠错码，该技术是在空域中进行的。为了了解本文所提出的在CWT域上所提出的算法，我们在DWT域上采用同样的嵌入算法来和它进行比较。最后，我们要通过试验来检测两种算法对于选择性攻击(比如JPEG，JPEG2000压缩)的鲁棒性，并利用水印检测软件基准来对他们进行比较。     

Acoustic feature selection for automatic emotion recognition from speech

Emotional expression and understanding are normal instincts of human beings, but automatical emotion recognition from speech without referring any language or linguistic information remains an unclosed problem. The limited size of existing emotional data samples, and the relative higher dimensionality have outstripped many dimensionality reduction and feature selection algorithms. This paper focuses on the data preprocessing techniques which aim to extract the most effective acoustic features to improve the performance of the emotion recognition. A novel algorithm is presented in this paper, which can be applied on a small sized data set with a high number of features. The presented algorithm integrates the advantages from a decision tree method and the random forest ensemble. Experiment results on a series of Chinese emotional speech data sets indicate that the presented algorithm can achieve improved results on emotional recognition, and outperform the commonly used Principle Component Analysis (PCA)/Multi-Dimensional Scaling (MDS) methods, and the more recently developed ISOMap dimensionality reduction method.     

基于小波变换的视频录播图像预处理及应用

针对传统图像预处理中图像信息损失的问题,提出基于小波变换的图像去噪和增强的算法。实验证明基于小波变换的图像预处理方法能在去噪和增强的同时,保留了图像在时间和空间域的信息,为视频录播后续的目标检测与跟踪提供了高质量图像,提高了录播系统的跟踪准确性。     

基于最小二乘法的行波启动元件新算法

文章提出了利用小波变换进行故障暂态信号奇异性检测的新方法,分析了目前的基于Lipschitz系数的行波启动元件算法,指出其最主要的缺点是只利用相邻尺度的WTMM或WTMS比值来确定Lipschitz系数。而通过拟合各尺度的数据,采用基于最小二乘法的Lipschitz系数进行估算。ATP-EMTP仿真结果表明,改进算法能够更精确的分析故障信号的奇异性。