Face Hallucination with Weighted Nuclear Norm Constraint

Face hallucination via patch-pairs leaning based methods has been wildly used in the past several years. Some position-patch based face hallucination methods have been proposed to improve the representation power of image patch and obtain the optimal regressive weighted vector. The rationale behind the position-patch based face hallucination is the fact that human face is always highly structured and consequently positioned and it plays an increasingly important role in the reconstruction. However, in the existing position-patch based methods, the probe image patch is usually represented as a linear combination of the corresponding patches of some training images, and the reconstruction residual is usually measured using the vector norm such as 1-norm and 2-norm. Since the vector norms neglect two-dimensional structures inside the residual, the final reconstruction performance is not very satisfactory. To cope with this problem, we present a weighted nuclear-norm constrained sparse coding (WNCSC) model for position-patch based face hallucination. In addition, an efficient algorithm for the WNCSC is developed using the alternating direction method of multipliers (ADMM) and the method of augmented Lagrange multipliers (ALM). The advantages of the proposed model are twofold: in order to fully make use of low-rank structure information of the reconstruction residual, the weighted nuclear norm is applied to measure the residual matrix, which is able to alleviate the bias between input patches and training data, and it is more robust than the Euclidean distance (2-norm); the more flexible selection method for rank components can determine the optimal combination weights and adaptively choose the relevant and nearest hallucinated neighbors. Finally, experimental results prove that the proposed method outperforms the related state-of-the-art methods in both quantitative and visual comparisons.     

Gradient-Based Low Rank Method for Highly Undersampled Magnetic Resonance Imaging Reconstruction

Recently, exploiting low rank property of the data accomplished by the non-convex optimization has shown great potential to decrease measurements for compressed sensing. In this paper, the low rank regularization is adopted to gradient similarity minimization, and applied for highly undersampled magnetic resonance imaging (MRI) reconstruction, termed gradient-based low rank MRI reconstruction (GLRMRI). In the proposed method, by incorporating the spatially adaptive iterative singular-value thresholding (SAIST) to optimize our gradient scheme, the deterministic annealing iterates the procedure efficiently and superior reconstruction performance is achieved. Extensive experimental results have consistently demonstrated that GLRMRI recovers both realvalued MR images and complex-valued MR data accurately, especially in the edge preserving perspective, and outperforms the current state-of-the-art approaches in terms of higher peak signal to noise ratio (PSNR) and lower high-frequency error norm (HFEN) values.     

基于图正则化和Schatten-p范数最小化的交通数据恢复

为充分利用交通数据低秩特性与局部近邻关系，准确恢复交通数据采集系统中的缺失数据，首先，应用基于核范数的低秩矩阵补全模型对交通数据矩阵进行预插补，以获得缺失值的初始估计，基于此，构建表征数据局部近邻结构的图模型；然后，提出融合图正则化和Schatten-p范数最小化的交通数据缺失值恢复模型；进一步，提出基于交替方向乘子框架的优化算法，求解缺失值恢复的最优化问题，得到最终的数据恢复结果；最后，用实际的高速公路交通流量和速度数据比较多种方法的恢复误差，同时给出所提方法的参数敏感性分析. 实验结果表明:在完全随机缺失、随机缺失和混合缺失模式下，缺失率为10% ~ 50%时，相比于局部最小二乘、概率主成分分析和低秩矩阵补全等方法，基于图正则化和Schatten-p范数最小化的算法恢复误差降低了3.02% ~ 28.49%.      

基于秩最小化矩阵去噪的船舶轨迹重构方法

针对船舶自动识别系统(Automatic Identification System，AIS)在实际应用中存在错误数据频发、数据丢包等问题，本文提出一种基于秩最小化矩阵去噪的船舶轨迹重构方法，利用去噪实现轨迹重构，同时，实现对轨迹的去噪和缺失补全。该方法通过线性插值实现经度对齐，将轨迹数据转化为轨迹矩阵，从而补全轨迹中的缺失值。由于补全结果存在非常大的误差，因此，引入 PLR(Patch-Based Low-Rank Minimization)算法去噪，消除误差。同时，为进一步提升补全效果，通 过2D-VMD(Two-Dimensional Variational Mode Decomposition)算法将矩阵分解为不同频率的IMF (Intrinsic Mode Function)，并分别进行PLR去噪，合并去噪结果，得到最终重构后轨迹。本文以长江武汉段水域船舶AIS轨迹为研究对象，通过实验证明该方法在不同缺失比例以及随机缺失和连续缺失两种情境下具有鲁棒性和较强的稳定性；并与 HALRTC(High-Accuracy Low-Rank Tensor Completion)、TRMF(Temporal Regularized Matrix Factorization)等方法进行比较，结果表明， 该方法相较于HALRTC等方法具有更高的精度，并在高损失率下表现出较好的重构效果。     

Fault diagnosis for wind turbine based on improved extreme learning machine

A fault diagnosis method based on improved extreme learning machine (IELM) is proposed to solve the weakness (weak generalization ability, low diagnostic rate) of traditional fault diagnosis with feedforward neural network algorithm. This method fuses signal feature vectors, extracts six parameters as the principal component analysis (PCA) variables, and calculates correlation coefficient matrix among the variables. The weight values of control parameters in the extreme learning model are dynamically adjusted according to the test samples’ constantly changing. Consequently, the weight fixed drawback in the original model can be remedied. A fault simulation experiment platform for wind turbine drive system is built, eight kinds of fault modes are diagnosed by the improved extreme learning model, and the result is compared with that of other machine learning methods. The experiment indicates that the method can enhance the accuracy and generalization ability of diagnosis, and increase the computing speed. It is convenient for engineering application.     

含有非线性电阻的动态电路唯一稳态的范数判定方法

以矩阵范数为工具，得到了确定含有非线性电阻的动态电路唯一稳态的条件，并给出了一个算例，其结果与理论分析吻合．本文的结果表明，含有非线性电阻的动态电路的唯一稳态．可以用矩阵范数决定．而且这个方法可以应用于任意阶的非线性电路．判据适用范围更广，结果便于应用，在理论与实用两个方面，都有重要意义．     

基于主成分分析法的沥青路面使用性能评价

针对沥青路面使用性能综合评价复杂性、模糊性及缺乏合理评价方法等问题,采用主成分分析法,选取现行规范确定的沥青路面使用性能评价指标,借助Matlab数学工具对沥青路面使用性能进行综合评价;并与已有文献中利用灰色聚类决策评价方法所得结果进行比较。分析表明:主成分分析法用于评价沥青路面使用性能是可行的,而且能客观地反映评价指标对综合评价值影响的强弱,具有特有的优越性。     

基于改进低秩矩阵补全的交通量数据缺失值插补方法

提出了一种低秩矩阵补全的改进方法以研究道路交通量数据缺失值插补问题。应用基于核范数的低秩矩阵补全对交通量数据矩阵中的缺失值进行第1轮插补; 通过层次聚类算法将交通量数据划分为不同类别, 使得同类中的数据具有较强相关性, 异类中的数据具有较弱的相关性; 在每类样本上应用低秩矩阵补全得到缺失值的第2轮插补; 为了减少聚类数的影响, 提出最小二乘回归集成学习方法将不同聚类数下的插补结果进行融合, 得到最终的交通量数据插补结果; 用美国俄勒冈州波特兰市的交通量数据比较了5种方法的插补误差, 并分析了不同聚类数和距离度量方法的影响。研究结果表明: 在完全随机缺失模式下, 缺失率为10%~60%时, 其相对于传统的低秩矩阵补全模型的插补误差降低了5.93%~9.11%;在随机缺失和混合缺失模式下, 插补误差也分别降低了8.32%~9.55%和8.14%~9.20%;集成不同聚类数下的多个插补结果比单一聚类数下的插补误差降低2.62%~4.76%。可见, 在3种数据缺失模式下, 改进低秩矩阵补全方法降低了交通量数据的插补误差, 能有效提高插补后交通量数据的有效性。      

Pulsed eddy current signal denoising based on singular value decomposition

The noise as an undesired phenomenon often appears in the pulsed eddy current testing (PECT) signal, and it is difficult to recognize the character of the testing signal. One of the most common noises presented in the PECT signal is the Gaussian noise, since it is caused by the testing environment. A new denoising approach based on singular value decomposition (SVD) is proposed in this paper to reduce the Gaussian noise of PECT signal. The approach first discusses the relationship between signal to noise ratio (SNR) and negentropy of PECT signal. Then the Hankel matrix of PECT signal is constructed for noise reduction, and the matrix is divided into noise subspace and signal subspace by a singular valve threshold. Based on the theory of negentropy, the optimal matrix dimension and threshold are chosen to improve the performance of denoising. The denoised signal Hankel matrix is reconstructed by the singular values of signal subspace, and the denoised signal is finally extracted from this matrix. Experiment is performed to verify the feasibility of the proposed approach, and the results indicate that the proposed approach can reduce the Gaussian noise of PECT signal more effectively compared with other existing approaches.     

离散小波变换快速算法中预滤波器的选取方法研究

离散小波变换快速算法使得小波变换在信号处理领域得到广泛应用,其中抽样空间上测量值的求取是快速算法中的重要步骤。在论述离散小波变换快速算法的基础上,研究了对抽样空间上测量值的求取方法,即预滤波器的小波选取法,直接选取法,取样函数法及其和特点,探讨了一种基于小波系数范数误差极小的预滤波器优化设计方法,仿真分析各种方法的误差,结果表明优化方法具有较好的分解精度。     

基于二维主分量分析的人耳身份识别研究

有效的特征提取方法是解决人耳身份识别任务的关键之一。以主分量分析（PCA）为代表的线性子空间方法在特征提取工作中得到了广泛应用。为了更有效地提取人耳图像特征并减少运算量,将基于二维图像矩阵的2D-PCA方法应用于人耳身份识别。针对三个USTB人耳图像库,采用最近邻分类器,研究了选用不同的特征维数、贡献率,及不同的相似性测度时,2D-PCA方法与传统的PCA方法的识别性能。交叉验证的实验结果表明：2D-PCA方法较PCA方法获得了更短的训练时间和更高的识别率,说明基于图像矩阵的2D—PCA方法是一种效率更高,鲁棒性更强的人耳身份识别方法。     

A New Robust Image Matching Method Based on Distance Reciprocal

Object matching between two-dimensional images is an important problem in computer vision. The purpose of object matching is to decide the similarity between two objects. A new robust image matching method based on distance reciprocal was presented. The distance reciprocal is based on human visual perception. This method is simple and effective. Moreover, it is robust against noise. The experiments show that this method outperforms the Hausdorff distance, when the images with noise interfered need to be recognized.     

PASSIVE CONTROL FOR A CLASS OF UNCERTAIN TIME-DELAY SYSTEMS

Introduction　　 Control of time- delay systems has been an at-tractive field in control theory and applicationssince the time- delay systems are frequently en-countered in the real world.They are much differ-ent from their non- delay counterparts.For exam-ple,it is well known that the existence of delaysmay degrade the respond of the closed- loop sys-tems,or even induce instability[1] .Hence,theavailable results of non- delay sytems can notbe di-rectly applied to time- delay systems in genera…     

复杂生产过程的小故障检测与分类方法

为监测复杂生产过程的状态,根据多元统计过程控制方法和支持向量机理论,将累积和控制图原理扩展为多变量的形式对过程数据进行预处理,并通过主元分析方法提取复杂生产过程的关键信息,得到有效的小故障数据,进而构建计算正常数据的统计量阀值及故障数据的Hotelling T平方统计值（T2）和平方预测误差值,实现了复杂生产过程的小故障模式检测,并采用支持向量机多分类方法将检测到的故障进行了分类.沥青混合料生产过程的仿真研究表明:在集料均值发生小波动、周期性上升和下降3种小故障模式下,故障检测识别率均达到95%,与主元分析方法相比平均提高了75%;分类准确率达到92.5%,与BP神经网络方法相比提高了19.3%.      

基于马尔科夫链的恶劣天气船舶行为预测

准确预测恶劣天气时船舶行为是港口水域应急调度和高效安全管理的关键,提出一种基于马尔科夫链的恶劣天气船舶行为预测模型.根据分析船舶行为实际变化规律,得出船舶行为是由船舶在不同空间进行状态转移而形成,然后结合船舶行为分布数据采用经验风险最小化策略求解状态转移矩阵,最后通过初始状态分布和状态转移矩阵预测恶劣天气时船舶行为.实例验证和分析结果表明该预测模型能科学准确地预测恶劣天气时船舶行为,可为港口应急调度与高效管理提供有效参考.     

基于计划行为理论的竞争驾驶行为建模0

为获取竞争驾驶行为的潜变量因子,分析驾驶员竞争驾驶行为的产生动机.对 225 名驾驶员进行了网上问卷调查,通过主成分分析和验证性因子分析提取竞争驾驶行 为的4 个潜变量因子：速度领先、空间占用、路权争夺和空间领先争夺.基于计划行为理 论,构建了竞争驾驶意图与行为关系结构方程模型.研究表明,驾驶员的竞争驾驶态度、主 观标准、知觉行为控制等心理因素和社会环境外界因素,通过行为意向能够很好地对竞 争驾驶行为进行预测.社会环境因素对竞争驾驶意图有显著影响,为驾驶员竞争驾驶行为 的矫正提供了有效途径.     

基于变分模态分解和奇异值分解的结构模态参数识别方法

为了准确获得结构的固有频率、阻尼比与振型, 将变分模态分解与奇异值分解相结合, 提出一种新的结构模态参数识别方法; 基于已有时频参数识别方法, 根据测量的脉冲激励与加速度响应估计系统的频响函数, 对系统的频响函数进行反傅里叶变换得到脉冲响应函数; 对各测点的脉冲响应函数进行变分模态分解, 得到与结构固有频率对应的本征模态分量; 提取本征模态分量的固有频率, 利用与固有频率相近的本征模态分量作为行向量构造奇异值分解矩阵, 对所构矩阵做奇异值分解, 利用最大奇异值重构左、右奇异值向量, 识别结构的振型、固有频率和阻尼比; 通过四自由度质量-弹簧-阻尼模态仿真试验和车体横梁锤击模态试验, 验证了所提出的模态参数识别方法的有效性。研究结果表明: 在四自由度理论模型参数识别中, 系统固有频率和阻尼比的识别结果与理论计算结果的最大相对误差分别不超过0.025%和1.490%, 理论计算与识别的1~4阶振型的模态置信度分别为0.999、1.000、0.999和0.999;在车体横梁锤击模态试验中, 提出方法识别的固有频率和阻尼比与理论计算结果的最大相对误差分别不超过1.57%和1.47%, 且车体横梁的理论振型与识别振型趋势相同。可见, 提出的方法能有效识别结构的模态参数。      

基于奇异值分解技术的轮边减速器故障信号降噪研究

利用改进的奇异值分解技术,用仿真信号验证了该技术对轮边减速器齿轮故障特征提取的有效性,并从模拟信号中提取出了故障特征频率。研究发现,噪声的奇异值分布趋于直线,凸显出了有用信号的奇异值,有利于特征信号提取。仿真结果表明,该方法能在强噪声背景下提取行星系统齿轮故障特征,为轮边减速器故障诊断提供了一个新的思路。     

基于正交变换的区间Ⅱ型模糊模型结构精简

针对奇异值-QR(SVD-QR)分解方法存在有效奇异值难以确定的问题,提出采用列选主QR分解方法对模糊模型结构进行分析.运用该方法分析从模糊模型抽取的2个激活强度矩阵,利用矩阵R主对角元素作为判断规则重要性的依据,根据矩阵Π中每列值为1的元素位置确定所对应的规则,从而选取重要的规则,构建简约的区间Ⅱ型模糊模型.将本文方法和奇异值-QR分解方法应用于混沌时间序列预测,同时还对比了两种方法选取的重要规则在不同样本条件下的适应能力.结果表明,两种方法选取的重要规则存在明显差异,并且采用本文方法可以获得更小的误差,平均误差为0.108 6;在不同样本条件下采用本文方法所得误差基本一致,具有更强的泛化能力.      

多跨先简支后连续钢筋混凝土空心板桥梁冲击系数研究

通过对动载试验实测值与基于跨长、频率两类规范计算值比较,以及对相关影响因素分析,表明多跨先简支后连续钢筋混凝土空心板桥梁冲击系数,与频率相关,但不遵守规范规定的单调函数规律,满足一定的域值,与动荷载速度相关,满足高阶多项式函数关系,与桥梁孔位基本无关。