Patch-based vehicle logo detection with patch intensity and weight matrix

A patch-based method for detecting vehicle logos using prior knowledge is proposed.By representing the coarse region of the logo with the weight matrix of patch intensity and position,the proposed method is robust to bad and complex environmental conditions.The bounding-box of the logo is extracted by a thershloding approach.Experimental results show that 93.58% location accuracy is achieved with 1100 images under various environmental conditions,indicating that the proposed method is effective and suitable for the location of vehicle logo in practical applications.

Spatial weight matrix in dimensionality reduction reconstruction for microelectromechanical system-based photoacoustic microscopy

A micro-electromechanical system(MEMS)scanning mirror accelerates the raster scanning of optical-resolution photoacoustic microscopy(OR-PAM).However,the nonlinear tilt angular-voltage characteristic of a MEMS mirror introduces distortion into the maximum back-projection image.Moreover,the size of the airy disk,ultrasonic sensor properties,and thermal effects decrease the resolution.Thus,in this study,we proposed a spatial weight matrix(SWM)with a dimensionality reduction for image reconstruction.The three-layer SWM contains the invariable information of the system,which includes a spatial dependent distortion correction and 3D deconvolution.We employed an ordinal-valued Markov random field and the Harris Stephen algorithm,as well as a modified delay-and-sum method during a time reversal.The results from the experiments and a quantitative analysis demonstrate that images can be effectively reconstructed using an SWM;this is also true for severely distorted images.The index of the mutual information between the reference images and registered images was 70.33 times higher than the initial index,on average.Moreover,the peak signal-to-noise ratio was increased by 17.08%after 3D deconvolution.This accomplishment offers a practical approach to image reconstruction and a promising method to achieve a real-time distortion correction for MEMS-based OR-PAM.

Drawing Weighted Directed Graph from It's Adjacency Matrix

This paper proposes an algorithm for building weighted directed graph, defmes the weighted directed relationship matrix of the graph, and describes algorithm implementation using this matrix. Based on this algorithm, an effective way for building and drawing weighted directed graphs is presented, forming a foundation for visual implementation of the algorithm in the graph theory.

Optimal strategy of searching FPD weights scanning matrix using GA-PSO

This paper discusses a kind of optimal method used for searching flat panel display （FPD） scanning matrix. The method adopts bionic algorithm： genetic algorithm （GA） and particle swarm optimization （PSO） algorithm. The method using single GA is more time-consuming, and the search efficiency is low in later evolution; the PSO algorithm is easily falling into the local optimal solution and appears the premature convergent phenomenon. Hence, a hybrid approach of GAPSO is found to optimize the search for high grayscale weights scanning matrix. Finally in the acceptable time, it finds a weight scanning matrix （WSM） of 256 gray scales with Matlab, whose scanning efficiency reaches 94.73% and the linearity is very good.

Research on consistency measurement and weight estimation approach of hybrid uncertain comparison matrix

The consistency measurement and weight estimation approach of the hybrid uncertain comparison matrix in the analytic hierarchy process （AHP） are studied. First, the decision-making satisfaction membership function is defined based on the decision making＇s allowable error. Then, the weight model based on the maximal satisfactory consistency idea is suggested, and the consistency index is put forward. Moreover, the weight distributing value model is developed to solve the decision making misleading problem since the multioptimization solutions in the former model. Finally, the weights are ranked based on the possibility degree approach to obtain the ultimate order.

Wide-Area Delay-Dependent Adaptive Supervisory Control of Multi-machine Power System Based on Improve Free Weighting Matrix Approach

The paper demonstrates the possibility to enhance the damping of inter-area oscillations using Wide Area Measurement (WAM) based adaptive supervisory controller (ASC) which considers the wide-area signal transmission delays. The paper uses an LMI-based iterative nonlinear optimization algorithm to establish a method of designing state-feedback controllers for power systems with a time-varying delay. This method is based on the delay-dependent stabilization conditions obtained by the improved free weighting matrix (IFWM) approach. In the stabilization conditions, the upper bound of feedback signal’s transmission delays is taken into consideration. Combining theoriesof state feedback control and state observer, the ASC is designed and time-delay output feedback robust controller is realized for power system. The ASC uses the input information from Phase Measurement Units (PMUs) in the system and dispatches supplementary control signals to the available local controllers. The design of the ASC is explained in detail and its performance validated by time domain simulations on a New England test power system (NETPS).

基于Matrix Profile的时间序列分割技术改进

时间序列分割是数据挖掘领域中的一个重要研究方向.目前基于矩阵轮廓(matrix profile,MP)的时间序列分割技术得到了越来越多研究人员的关注,并且取得了不错的研究成果.不过该技术及其衍生算法仍然存在不足:首先,基于矩阵轮廓的快速低代价语义分割算法中对给定活动状态的时间序列分割时,最近邻之间通过弧进行连接,会出现弧跨越非目标活动状态匹配相似子序列问题;其次,现有提取分割点算法在提取分割点时采用给定长度窗口,容易得到与真实值偏差较大的分割点,降低准确性.针对以上问题,提出一种限制弧跨越的时间序列分割算法(limit arc curve cross-FLOSS,LAC-FLOSS),该算法给弧添加权重,形成一种带权弧,并通过设置匹配距离阈值解决弧的跨状态子序列误匹配问题.此外,提出一种改进的提取分割点算法(improved extract regimes,IER),它通过纠正弧跨越(corrected arc crossings,CAC)序列的形状特性,从波谷中提取极值,避免直接使用窗口在非拐点处取到分割点的问题.在公开数据集datasets_seg和MobiAct上面进行对比实验,验证以上两种解决方案的可行性和有效性.

A Note on the Spectral Radius of Weighted Signless Laplacian Matrix

A weighted graph is a graph that has a numeric label associated with each edge, called the weight of edge. In many applications, the edge weights are usually represented by nonnegative integers or square matrices. The weighted signless Laplacian matrix of a weighted graph is defined as the sum of adjacency matrix and degree matrix of same weighted graph. In this paper, a brief overview of the notation and concepts of weighted graphs that will be used throughout this study is given. In Section 2, the weighted signless Laplacian matrix of simple connected weighted graphs is considered, some upper bounds for the spectral radius of the weighted signless Laplacian matrix are obtained and some results on weighted and unweighted graphs are found.

企业研发经费的就业效应——基于制造业资本品上下游行业的空间计量分析

加大企业研发经费(R&D经费)投入会对就业产生何种影响是非常值得关注的问题。本文首先建立了基于研发而形成的资本品上下游行业模型,从理论上分析企业R&D经费对就业的影响。然后,基于我国2012—2021年制造业行业数据,运用空间计量的方法对企业R&D经费的就业规模总效应和上下游行业就业结构变化进行实证检验。研究发现:我国企业R&D经费对就业规模总效应不显著;但资本品上游行业企业R&D经费增加会负向影响下游行业就业,使上游行业就业占比增加,发生就业结构上迁。异质性分析表明,设备型上游行业组的就业结构上迁效应明显。机制分析表明,上游行业产品创新增加促进了下游行业的资本品使用,对下游行业劳动力产生替代,促使就业结构向上迁移。研究结论对优化创新和就业政策、减缓摩擦性失业有一定现实意义和参考价值。

基于一致性的赋权方法在课程过程性评价中的应用

针对课程平时成绩的权重比例,构建采用客观赋权法确定权重的动态代数模型.基于平时成绩的各项得分,分析项目之间的一致度,应用矩阵理论对权重进行一致性检验,给出一致性最强的权重比例.结合实例计算最优权重.

关于赋权非正则图的A_(α)特征值和特征向量

设G_(ω)=(G,ω)是一个赋权图,其邻接矩阵和赋权度对角矩阵分别A(G_(ω))和D(G_(ω))。对于α∈[0,1],G_(ω)的A_(α)-矩阵为A_(α)(G_(ω))=αD(G_(ω))+(1-α)A(G_(ω))。对于连通赋权非正则图G_(ω),给出了其关于A_(α)-特征值的一些界,并得到了A_(α)-谱半径对应的特征向量中最大分量与最小分量比值的下界。

基于加权非负矩阵分解的异常声音检测方法研究

异常声音检测方法多用强标签数据进行训练,而高质量的强标签音频数据标注难度较大、收集成本高昂。针对现有异常音频检测方法使用弱标签数据会受到非平稳和时变噪声的干扰,导致训练结果较差、准确率低的问题,提出一种基于音频频谱的加权非负矩阵分解WNMF方法。该方法使用WNMF对弱标签和无标签数据进行标记,并分离目标声音事件和背景噪声。在适当的权值下,WNMF改变标记时不同频段音频信息的重要性,以抑制噪声,提高分离质量,使其逼近全监督模型训练的效果;之后使用卷积神经网络产生帧级预测和音频标签预测。仿真实验结果表明,该方法的准确率相比于传统NMF处理弱标签数据的方法提升了4.8%。

基于Relief算法的不平衡数据分类分级算法仿真

不平衡数据的分类分级是保证大数据技术高效使用过程中不可缺少的环节,但分类分级过程易受数据属性、冗余性、不均衡性等问题的干扰。为解决上述问题,提出不平衡数据朴素贝叶斯分类分级算法。采用合成少数类过采样技术降低数据的不平衡度,通过距离相关系数与最大信息系数完成不平衡数据的特征选择与筛选,采用Relief算法对筛选的特征做权重分配,并输入到朴素贝叶斯模型中实现分类,再结合动态阈值算法完成数据的分级。实验结果表明,所提算法的运行时间短、分类精度高,能够有效提升数据处理效果。

基于现场可编程门阵列的矩阵求逆算法设计

由于自适应抗干扰算法在更新最优权值时存在时间延时,导致很难满足动态环境下的权值更新率要求。针对该情况已有学者对如何实现快速采样矩阵求逆算法进行研究,但仍存在只适用于低维矩阵,且权值更新率慢的问题。为解决上述问题,提出了一种基于Cholesky分解的采样矩阵求逆算法实现架构。该实现架构主要包括协方差矩阵计算模块、Cholesky分解模块、计算下三角矩阵L的逆矩阵模块、三角矩阵相乘和权值计算模块。设计采用流水线加状态机实现结构有效地解决了因高阶采样矩阵求逆运算量大产生的权值更新率慢的问题。仿真结果表明,在现场可编程门阵列(field programmable gate array,FPGA)的硬件平台上,对于56阶采样矩阵,在100 MHz工作频率下,一次权值的更新时间仅需要1.2 ms。本文所提的实现架构为自适应抗干扰快速求解权值提供了一种切实可行的解决方案,对存在类似需求的权值求解系统具有一定的参考价值。

基于TFAHP的电炉企业多维一体化财务模式评价模型

电炉企业的财务由于其参数指标众多,对相关模式评价过程不能运用单一阈值理论,评价过程一直是个难题。设计基于TFAHP的电炉企业多维一体化财务模式评价模型。通过微观模拟电炉企业财务工资过程,确定评价指标,由此构建模糊评价矩阵,依据模糊评价矩阵,构建二级评价矩阵,在单一准则下计算各评价指标的相对权重,采用三角模糊数层次分析法确定总权重,由此构建评价模型,获取评价结果,设定评价等级集合及区间值,给出具体评语等级。实验可知,所构建模型进行多维一体化财务模式评价在电炉企业应用效果较好,评价效率较高。

空间计量经济学前沿理论、方法与应用研究综述

数字经济时代区域要素流动愈发明显,经济活动的空间组织和经济地理正发生着剧烈的、持续的和趋势性的变化,对空间经济理论与实证研究提出了重大需求,在此背景下空间计量经济学模型的应用价值也越来越为重要。为深化空间计量经济学模型的理论认识,促进其在经济管理学领域实证研究中的规范应用,文章在充分收集文献资料的基础上,着眼于空间计量经济学前沿理论、方法与应用,从空间计量经济学理论及技术、模型体系及拓展、国内应用现状、不足及反思等方面进行了较为系统的文献分析与总结,并展望了其发展前景和方向。未来国内空间计量经济学应在技术理论研究、模型应用体系,及方向上进行不断推进,以提升空间计量经济学理论与应用研究水平。

基于图的自适应加权多视图聚类

针对现有的基于图的多视图聚类算法没有考虑不同视图的权重和视图数据存在噪声的问题,提出一种基于图的自适应加权多视图聚类算法。通过自适应邻域学习从原始数据中构造多个关系图,引入视图权重调节参数,减少噪声的影响;通过自适应学习将各个关系图融合成统一关系图,通过秩约束优化使数据点自动划分成所需的簇,从而得到聚类结果。在多视图数据集上的实验结果表明了该算法的有效性。

各向异性的L_(0)正则化图像平滑方法

现有的图像平滑方法缺乏灵活性,会导致边缘不清晰、结构缺失和过度锐化等问题。文中提出一种新的自适应加权矩阵的正则化方法,主要应用于图像平滑,并且可以扩展到其他应用。提出的模型设计了一个新的正则化项,基于梯度算子▽和自适应加权矩阵T组合为L_(0)范数正则化项,使得模型具有各向异性。通过为不同梯度方向赋予不同的权重,以此来刻画平滑图像的局部结构,更好地展现局部特征,防止过度平滑。由于所提出的模型是非光滑且非凸的,在求解上比较复杂,因此采用ADMM算法对模型进行求解。把目标函数分解成几个易求解的子问题,分别对每个子问题求解,最终得到模型的最优解。主客观实验表明,提出的模型在视觉效果以及数值方面都有明显的提高。

宽度学习系统中鲁棒性权值矩阵组合的筛选方法

宽度学习系统(BLS)具有出色的计算效率和预测准确性;然而,在传统BLS框架中,权值矩阵采用随机生成的方式,存在学习结果不稳定的风险。因此,设计一种BLS中鲁棒性权值矩阵组合的筛选方法(RWS-BLS)。首先,通过4组函数数据的验证,揭示随机权值矩阵在样本整体训练误差上的显著差异性;其次,研究权值矩阵组合的形式,放宽筛选条件的严格最优限制,将最优转换为较优,并将误差最小值限定在指定范围内,定义精英组合等条件;最后,得到可靠的权值矩阵的组合,有效降低随机性影响,并建立稳健的模型。实验结果表明,在16组模拟数据、NORB数据集和5组UCI回归数据集上,在数据更换或受噪声扰动的情况下,与BLS方法相比,所提方法的均方误差(MSE)下降了7.32%、8.73%和1.63%。RWS-BLS为BLS提供了一种模型平稳性研究的方向,提高了含有随机参数模型的效率和稳定性,并对涉及随机参数的其他机器学习方法具有借鉴作用。

基于不同权重的土地利用空间自相关模型对比分析——以元谋县为例

[目的]研究不同空间权重矩阵对土地利用空间自相关模型的影响。[方法]以元谋县土地利用变量为基础,首先分析不同权重矩阵下土地利用类型及其驱动因子的空间自相关性,建立基于queen权重矩阵、rook权重矩阵和距离阈值权重矩阵的耕地时空格局演变的空间自回归模型,选取拟合优度、最大似然对数值、赤池信息准则、施瓦茨信息准则、影响因子数量和模型残差空间自相关性等参数作为模型评价指标,对比分析经典线性回归模型、空间滞后模型和空间误差模型的差异。[结果]2018年元谋县土地利用类型和土地利用驱动因子在不同空间权重矩阵下均呈现出较强的空间正相关关系。通过对比3种权重矩阵的经典线性回归模型、空间滞后模型和空间误差模型,发现在同一空间自相关模型中,基于距离阈值权重矩阵的空间自回归模型的拟合效果较好,基于同一个空间权重矩阵,空间误差模型的拟合效果较好。[结论]空间相关性与空间权重矩阵有关,基于距离阈值矩阵的空间误差模型的拟合度最高,解释能力最强,能更好表征云南高原山区的土地利用格局的空间演变。