SVM在自主履带车辆转向问题上的仿真研究

SVM(支持向量机)是一种应用广泛的机器学习分类、回归和完成其他学习任务的方法。文章简要介绍了支持向量机的原理,并以一维数据为基础,采用基于-不敏感损失函数的-SVR与基于最小二乘基本原理的曲线拟合法,在MATLAB平台下,对多项式函数、包含指数函数的三角函数分别进行回归比较。最后对多维数据的某型自主履带车辆转向角度进行回归预测。结果表明支持向量机的方法具有应用范围广、逼近精度高、泛化性能好的特点。     

基于支撑向量机的人脸识别技术

文中提出了一种基于支撑向量机的人脸识别方法.该方法与传统方法相比,克服了后者固有的过学习和欠学习问题,并且对复杂模式的识别能力强,达到了很高的人脸识别率.在对训练图像进行预处理之后,使用主成分分析方法对人脸图像进行特征提取和选择,用所选择的人脸特征向量训练多个支撑向量机,最后用训练好的支撑向量机进行人脸识别.文中将支撑向量机性能和传统方法进行了对比,并且对不同核函数的支撑向量机的性能也进行了对比.发现当特征脸数量不同时,不同核函数支撑向量机的性能也不同.总体而言,二阶多项式支撑向量机在人脸识别问题中具有更好的性能.     

基于支撑向量机的人脸识别技术

文中提出了一种基于支撑向量的人脸识别方法。该方法与传统方法相比,克服了后者固有的过学习和欠学习问题,并且对复杂模式的能力强,达到了很高的人脸识识别率。在对训练图像进行预处理之后,使用主成分析方法对人脸图像进行特征提取和选择,用所选择的人脸特征向最训练多个支撑向量机,最后用训练好的支撑向量机进行人脸识别。文中将支撑向量机性能和传统方法进行了对比,并且对不同核函数的支撑向量机的性能也进行了对比。发现当特征脸数量不同时,不同核函数支撑向量机的性能也不同。总体而言,二阶多项式支撑向量机在人脸识别问题中具有更好的性能。     

基于ε-支持向量回归机的盲均衡算法

在研究支持向量机的基础上,针对其优良的泛化性能,提出了基于ε-支持向量回归机(ε-SVR)的盲均衡算法.采用迭代加权二次规划方法进行求解,给出理论模型和算法推导,通过核函数推广到非线性盲均衡.计算机仿真结果表明,该算法在基带条件下对具有严重码间干扰的接收信号能够均衡处理,实现电话信道的零误差传输,相对于经典CMA算法而言,新算法在提高收敛性能的同时,提高收敛速度50%左右.尤其是在小样本情况下,计算复杂度低,收敛速度快.     

一种小波核函数的支持向量回归算法及应用

借助谐波小波函数在分析窄带信号方面的性能,利用支持向量回归算法,提出了一种基于谐波小波核函数和支持向量机相结合的谐波小波核支持向量回归算法,实现了小样本情况下微弱信号的精确检测.仿真和实测检测噪声数据的分析表明该算法可以有效地检测出舰船噪声中的线谱信号.     

多项式光滑孪生支持向量回归机

针对光滑孪生支持向量回归机（Smooth Twin Support Vector Regression ,STSVR）中Sigmoid函数逼近精度不高的问题,将正号函数展开为无穷多项式级数,由此得到一族光滑函数。采用该多项式光滑函数逼近孪生支持向量回归机的不可微项,并用New ton-Armijo算法求解相应的模型,提出了多项式光滑孪生支持向量回归机（Poly-nomial Smooth Twin Support Vector Regression ,PSTSVR）。不仅从理论上证明了PSTSVR的收敛性和满足任意阶光滑的性能,而且在人工数据集和UCI数据集上的实验表明了PSTSVR比STSVR具有更好的回归性能。     

基于分段核函数的支持向量机及其应用

支持向量机兼顾训练误差和推广性能,已受到机器学习领域的高度重视,而核函数的性能是支持向量机研究中的关键问题。研究了几种常见核函数对支持向量机推广性能的影响,并利用全局核函数和局部核函数的性质,提出了一种新的分段核函数的支持向量机。数据集上的仿真结果表明,该核函数对应的支持向量机泛化能力优于传统核函数对应的支持向量机,具有较好的预测性能。     

一种新的多类模式识别方法

提出了一种基于支撑向量机的多类分类器,用N-1个支撑向量机组合构成一个具有二叉树结构形式的N-多类分类器．讨论了该多类分类器的泛化推广能力,同时还提出了该多类分类器的基于特征空间的BTSVM学习算法,BTSVM算法使用核函数转换的方式计算特征空间的样本距离;采用类间最小距离最大化作为聚类准则,在每个决策结点产生两个最优子集;然后采用支撑向量机学习算法学习两个最优子集,确定决策结点的最优分类面．理论和实验结果表明,本文提出的基于支撑向量机的多类分类器在整体性能上要优于其它类似的分类器系统。     

记忆非线性功放预失真器SVM建模的算法改进

为提高支持向量机(SVM)ε不敏感损失函数下的回归算法的训练速度,提出了一种新的管道压缩模型,利用大ε值下的回归函数来预测小ε值下回归函数的支持向量.由该模型导出一种新的算法结构.在对记忆非线性功率放大器的SVM预失真器进行建模仿真中,将Keerthy的SMO算法同新的算法结构相结合,结果表明了新算法结构在不损失SVM预失真器性能的基础上,显著地提高了训练速度.     

基于复合核支持向量回归机的多类分类算法

针对传统支持向量机(SVM)在解决多类分类问题时需要训练多个分类器、存在不可分区域等问题,研究了基于支持向量回归机的多类分类算法。利用回归思想求解多类分类问题,将分类样本作为回归输入,样本的类别标识作为回归输出,通过支持向量回归机训练拟合出各样本与其类别标识之间的函数关系。将待分类样本代入回归函数,对其输出取整后即可得到样本类别。该算法仅使用1个分类器,明显简化了分类过程。另外,引入复合核函数来提高支持向量回归机的性能。采用加州大学欧文分校(UCI)例题库中的多类分类问题进行仿真验证,并将改进算法与传统算法作对比,结果表明改进算法在分类速度和准确率上都有显著提高。     

一种基于SVR的数字图像水印算法

以回归型支持向量机为基础,提出一种彩色数字图像水印算法。在小波域内选取特征向量并获得支持向量机训练模型,进而利用该训练模型嵌入和提取水印信息。该算法以保证不可感知性和鲁棒性的良好平衡为前提,实现了水印的盲检测。实验仿真表明,该算法不仅具有较好的不可感知性,而且对叠加噪声、JPEG压缩、锐化、平滑滤波、对比度增强、剪切等常规处理具有较好的鲁棒性,其整体性能优于一般基于支持向量机的图像水印方案。     

A novel image watermarking scheme against desynchronization attacks by SVR revision

In digital image watermarking, the watermark’s vulnerability to desynchronization attacks has long been a difficult problem. On the basis of support vector regression (SVR) theory and local image characteristics, a novel image watermarking scheme against desynchronization attacks by SVR revision is proposed in this paper. First, some pixels are randomly selected and the sum and variance of their neighboring pixels are calculated; second, the sum and variance are regarded as the training features and the pixel values as the training objective; third, the appropriate kernel function is chosen and trained, a SVR training model will be obtained. Finally, the sum and variance of all pixels’ neighboring pixels are selected as input vectors, the actual output can be obtained by using the well-trained SVR, and the digital watermark can be recovered by judging the output vector. Experimental results show that the proposed scheme is invisible and robust against common signals processing such as median filtering, sharpening, noise adding, and JPEG compression, etc., and robust against desynchronization attacks such as rotation, translation, scaling, row or column removal, shearing, local random bend, etc.     

基于支持向量回归的Contourlet域盲水印算法

为进一步提高基于支持向量机(SVM,support vector machine)水印算法的性能,提出了基于支持向量回归(SVR,support vector regression)的Contourlet域盲水印算法。首先对宿主图像进行Contourlet分解,然后利用SVM建立图像尺度内的局部相关性模型,根据模型的预测结果自适应地嵌入水印。实验结果表明,所提出的算法不仅具有较好的不可感知性,而且对叠加噪声、JPEG压缩、锐化、平滑滤波和对比度增强等常规图像信号处理以及旋转、剪切等几何攻击均具有较好的鲁棒性,其性能明显优于基于SVM的空间域和小波域的水印算法。     

基于机动特征辅助的MFR状态预测方法

针对多功能雷达（Multi-Function Radar,MFR）状态预测方法存在的鲁棒性、预测正确率不佳的问题,提出一种基于机动特征辅助的MFR状态预测方法.该方法将载机机动信息与常规侦收参数共同作为预测特征集,一方面利用支持向量回归（Support Vector Regression,SVR）和侦收信号特征集,得到常规预测模型,另一方面通过SVR和机动特征集,得到MFR各个状态间的转变概率模型;然后利用D-S证据理论得到最终预测状态.实验结果表明,与SVR和LSR方法相比,平均预测精度分别提高了6.97%和7.2%,同时具备更优异的鲁棒性.此外,提出的预测方法通过进一步的拓展,可应用于机械设备、道路交通等领域.     

基于支持向量回归的图像超分辨率重建算法

文章提出一种新的基于支持向量回归(SVR)和稀疏表示的图像超分辨重建算法。SVR对输入数据有良好预测输出类别能力。图像统计表明,图像块可以从过完备字典中通过稀疏线性组合很好的表示。对一幅低分辨率输入图像,可以将图像超分辨问题视为在高分辨图像中估计其像素位置。与传统的支持向量回归方法相比,本文采用的特征是不同类型的图像块的稀疏表示。研究表明,稀疏表示作为特征对噪声有一定的鲁棒性。实验结果表明,本文方法与传统支持向量回归方法相比在图像重建质量上有一定的优势。     

基于Contourlet域虚拟树结构和FOA-SVR的自适应鲁棒数字水印算法

为了更好地在Contourlet域里自适应地选择水印的嵌入位置和嵌入强度,克服Contourlet变换最后一层低频子带没有被划分的缺陷,给出了一种Contourlet域虚拟树结构的具体构造方法.基于该构造方法,结合混沌序列,利用果蝇算法(FOA)优化支持向量回归机(SVR)的参数,提出一种自适应的鲁棒数字水印算法,实验结果表明该算法在有较强的鲁棒性和安全性的同时,也具有很好的保真度.     

An adaptive network traffic prediction approach for LDoS attacks detection

Low‐rate denial of service (LDoS) attacks reduce throughput and degrade quality of service (QoS) of network services by sending out attack packets with relatively low average rate. LDoS attack flows are difficult to detect from normal traffic since it has the property of low average rate. The research on network traffic analysis and modeling shows that network traffic measurement data are irregular nonlinear time series. To characterize and analyze network traffic between attack and non‐attack situations, the adaptive normal and abnormal ν‐support vector regression (ν‐SVR) prediction models are constructed on the basis of the reconstructed phase space. In this paper, the dimension of reconstructed phase space for ν‐SVR is optimized by Bayesian information criteria method, and the parameter in the radial basis function is adaptively adjusted by minimizing the within‐class distance and maximizing the between‐class distance in the feature space. The nonthreshold decision function is obtained through calculating the prediction error of adaptive normal and abnormal ν‐SVR prediction models, which is adopted to detect LDoS attacks. Experiments in NS‐2 environment show that the adaptive ν‐SVR prediction model can effectively predict the network traffic measurement time series, and the probability distribution of time series generated by the adaptive ν‐SVR prediction model is quite similar to that of the network traffic measurement data. Experiments also clearly demonstrate the superiority of the proposed approach in LDoS attacks detection.     

Support vector regression via MCMC within evidence framework

This paper proposes a novel approach, Markov Chain Monte Carlo (MCMC) sampling approximation, to deal with intractable high-dimension integral in the evidence framework applied to Support Vector Regression (SVR). Unlike traditional variational or mean field method, the proposed approach follows the idea of MCMC, firstly draws some samples from the posterior distribution on SVR??s weight vector, and then approximates the expected output integrals by finite sums. Experimental results show the proposed approach is feasible and robust to noise. It also shows the performance of proposed approach and Relevance Vector Machine (RVM) is comparable under the noise circumstances. They give better robustness compared to standard SVR.     

No-reference image quality assessment based on hybrid model

The aim of research on the no-reference image quality assessment problem is to design models that can predict the quality of distorted images consistently with human visual perception. Due to the little prior knowledge of the images, it is still a difficult problem. This paper proposes a computational algorithm based on hybrid model to automatically extract vision perception features from raw image patches. Convolutional neural network (CNN) and support vector regression (SVR) are combined for this purpose. In the hybrid model, the CNN is trained as an efficient feature extractor, and the SVR performs as the regression operator. Extensive experiments demonstrate very competitive quality prediction performance of the proposed method.     

Support Vector Regression Model for Millimeter Wave Transitions

In this paper, we introduce a new method, support vector regression (SVR) method, to model millimeter wave transitions. SVR is based on the structural risk minimization (SRM) principle, which leads to good generalization ability for regression problem. The SVR model can be electromagnetically developed with a set of training data and testing data which produced by the electromagnetic simulation. Two Ka-band millimeter wave transitions, i.e., waveguide to microstrip transition and coaxial to waveguide adapter, are used as examples to validate the method. Experimental results show that the developed SVR models have a good predictive ability, and they are useful for interactive CAD of millimeter wave transitions.