基于归一化相关矩的多分辨率遥感图象融合

多传感器数据融合技术已广泛应用于遥感图象处理方面 .针对遥感多光谱图象空间分辨率较低的问题 ,提出了一种基于归一化相关矩的多分辨率图象融合方法 .该方法首先对图象进行二维小波变换 ,然后根据所得到的高频小波系数的一阶、二阶统计特征来定义图象局部灰度相关矩 ,并以此作为图象融合测度来对遥感图象进行多分辨率特征融合 ,从而得到包含更多信息和有效特征的融合图象 .仿真结果表明 ,融合后的图象在保留多光谱信息和提高空间分辨率上均能获得较好的效果 ,因而可以更好地用于目标识别、分类等遥感图象处理方面     

基于信息融合团队一致法的车辆识别方法

介绍了一种基于信息融合的团队一致法的车辆识别方法.该方法把传感器组看成共同协作环境下不确定性问题的一队"专家",按一定的规则得到最终的综合决策.将该方法应用在车辆识别的实验中,实验结果表明了该方法的有效性.     

基于不确定性度量的多特征融合跟踪

提出了一种新的基于特征不确定性度量的多特征融合跟踪算法. 首先, 针对粒子滤波跟踪算法中特征鉴别能力较弱且粒子分布相对分散时容易造成目标丢失的事实, 本文定义了一种新的特征不确定度量方法, 该度量可以在线调整不同类型特征对跟踪结果的贡献. 同时, 针对乘性和加性特征融合跟踪算法方法中存在的缺陷, 提出了一种自适应的多特征融合方法, 融合的结果既突出了状态后验分布中目标真实状态对应的峰值, 又对噪声不敏感, 从而提高了目标跟踪的鲁棒性. 各种场景下的实验结果比较表明: 新的融合跟踪算法比单特征跟踪、 乘性融合跟踪和加性融合跟踪有着更好的稳定性和鲁棒性.     

基于改进PIC参数的证据理论融合算法效能评估

基于PIC参数提出了TPIC参数,并提出用TPIC参数对证据理论融合系统性能进行评估的方法.融合的目的是为了得到对目标更准确、确定的认识,用TPIC参数给出了多证据融合系统融合输出对正确目标确定度的度量方法,通过度量融合后对正确目标的确定度来实现融合性能的定量评估,解决了以往熵方法评估仅以不确定性作为标准而不考虑融合结果正确性的问题,并由仿真实验验证了方法的有效性.     

基于多传感器的水声定位精度研究

针对TDOA定位算法中几何不确定性对定位精度的负面影响,提出一种基于多传感器信息融合的水声定位算法;该算法首先对多传感器合理布阵,确定若该个TDOA定位子系统,再对每个定位子系统用TDOA算法求得目标位置信息,最后依据几何精度因子(GDOP)对得到的若干个目标位置信息进行加权融合,得到最终定位结果;利用软件仿真对算法可行性进行了验证,结果表明,与基本的五元阵定位系统相比,该算法有效消除了因几何不确定性引起的误差,大大提高了定位精度。     

基于PCA变换与小波变换的遥感图象融合

针对多光谱图象与全色图象的融合问题,提出了一种基于PCA变换和小波变换的遥感图象融合方法.新方法通过对多光谱图象作PCA变换,首先得到3个主分量;然后,利用小波变换融合方法融合多光谱图象的第1主分量与全色图象,并用融合后的图象替代多光谱图象的第1主分量;最后,作PCA反变换来得到新的多光谱图象.主观视觉效果分析和客观统计参数评价分析表明,新方法的性能优于PCA变换融合方法、IHS变换融合方法和小波变换融合方法,该方法不仅较大地增强了结果图象空间细节的表现能力,而且很好地保留了多光谱图象的光谱信息.     

基于严凹函数的粗糙集不确定性度量

通过语义分析,提出了一种拓展的粗糙集不确定性度量公理化定义;将香农熵函数推广到严凹函数,提出了一类以条件概率为自变量、基于严凹函数的粗糙集不确定性度量公式,它是严凹函数值的加权平均.在此基础上,得到一系列粗糙集不确定性度量方法.从严凹函数视角讨论了基于模糊熵的不确定性度量方法,发现现有多种能够用于度量粗糙集不确定性的模糊熵函数都是所提出方法的特殊情形.比较了粗糙度、改进粗糙度和所提出方法的区别和联系,最后设计了一些算例,比较了各种方法的异同,验证了基于严凹函数的粗糙集不确定性度量与粗糙集不确定性语义是一致的.     

一种新的位置数据融合方法的研究

给出了一新的基于团队一致法的多传感器位置数据融合方法，该方法按传感器队中的每个成员的测量不确定性，构造团队期望效用函数（或密度），并基于该期望效用函数求得位置参数估计，其优点是可消除失效传感器和测量值为野值的传感器的影响，本文给出了仿真结果。     

一种证据不确定性度量方法及其应用

证据的冲突是导致证据理论融合结果不理想的重要因素,对证据冲突的处理一直是证据理论中要解决的关键问题。通过对多义度、不一致度和非特异度进行线性组合,提出一种新的证据不确定性度量方法。新方法能够更全面地涵盖证据体中所包含的不一致和非特异性两类不确定信息,使得证据不确定性度量结果所包含的信息更为完整。在此基础上,采用指数函数构造权重,对证据体进行预处理,然后采用DST、PCR2和PCR5进行融合。算例结果表明融合结果合理,新的证据不确定性度量方法有效。     

基于不确定性的数据融合系统性能评估

提出了一种基于不确定性的数据融合系统性能评估方法。数据融合的目的是为了提高融合后的信息量，信息不确定度的降低就相当于信息量的增加，本文从广义信息论出发度量信息不确定度，通过比较融合前后系统信息的不确定度来度量融合系统的性能。给出多目标系统中计算信息不确定度归一化变化的方法，实现了融合系统性能的定量评估。     

犹豫模糊信息相似性测度与群一致性测度及群决策应用

针对目前关于犹豫模糊运算与测度的研究中存在的不足,首先给出犹豫模糊熵函数的定义,并将其作为犹豫模糊信息不确定性测度,进而提出犹豫模糊信息特征向量概念,以信息特征向量为出发点对犹豫模糊距离测度和相似性测度展开研究;为优化群决策过程,提出基于完全优先关系的群一致性测度概念并研究其性质;最后,提出基于相似性测度和群一致性测度的群决策方法并结合算例验证所提出方法的有效性.     

基于语言共识模型的电子商务信用风险评价方法

电子商务信用风险评价能够更好地维护市场规则并防范交易主体的合法权益.从语言评价信息的角度,利用多属性群决策方法对电子商务信用风险评价方法进行探讨.首先,提出个体语言共识测度和群体语言共识测度;然后,针对共识性水平较低的决策群体,构建一种整数规划模型,用于调整决策者给出的初始语言决策信息;最后,提出一种基于语言共识模型的电子商务信用风险评价方法,并通过电子商务信用风险评价问题说明该方法的可行性和有效性.     

New approach for aggregating multi-sensory data

The task of sensory data fusion may involve the aggregation of sensory measurements that may be from different phenomenological domains and that, in many cases, could embrace some conflicting information cues. It is rather a challenge to find suitable strategies by which measurements obtained by the different sensors of the system can be aggregated so that a consistent interpretation of these measurements is achieved. In this article, we present a novel approach to achieve this goal. A recursive group utility function that is capable of bringing the group of sensors into consensus is used. After each sensor in the group gathers information relevant to the sensory task, the group engages in what we call the uncertainty estimation stage. This is an information theorybased process that allows each sensor to assess its self-uncertainty and the conditional uncertainties of other sensors. This process facilitates the computation of a weighting scheme that operates recursively on sensor observations until the group reaches a consensus. Whenever new observations are made, the uncertainty estimates of sensors are updated and used to compute a new weighting scheme. To demonstrate the efficacy and to show how the methodology works, the article discusses how the method can be used to tackle the multi-sensor object identification problem. © 1993 John Wiley & Sons, Inc.     

Linear uncertain extensions of the minimum cost consensus model based on uncertain distance and consensus utility

Uncertainty theory adopts the belief degree and uncertainty distribution to ensure good alignment with a decision-maker’s uncertain preferences, making the final decisions obtained from the consensus-reaching process closer to the actual decision-making scenarios. Under the constraints of the uncertain distance measure and consensus utility, this article explores the minimum-cost consensus model under various linear uncertainty distribution-based preferences. First, the uncertain distance is used to measure the deviation between individual opinions and the consensus through uncertainty distributions. A nonlinear analytical formula is derived to avoid the computational complexity of integral and piecewise function operations, thus reducing the calculation cost of the uncertain distance measure. The consensus utility function defined in this article characterizes the adjustment value and degree of aggregation of individual opinions. Three new consensus models are constructed based on the consensus utility and linear uncertainty distribution. The results show that, in complex group decision-making contexts, the uncertain consensus models are more flexible than traditional minimum-cost consensus models: compared with the high volatility of the adjusted opinions in traditional deterministic consensus models with crisp number-based preferences, the variation trends of both individual adjusted opinions and the collective opinion with a linear uncertainty distribution are much smoother and the fitting range is closer to reality. The introduction of the consensus utility not only reflects the relative changes of individual opinions, but also accounts for individual psychological changes during the opinion-adjustment process. Most importantly, it reduces the cost per unit of consensus utility, facilitates the determination of the optimal threshold for the consensus utility, and improves the efficiency of resource allocation.     

Adaptive consensus model with multiplicative linguistic preferences based on fuzzy information granulation

An adaptive consensus model based on fuzzy information granulation (fuzzy IG) is presented for group consensus decision-making problems with multiplicative linguistic preference relations (MLPRs). Firstly, a granular representation of linguistic terms is concerned with the triangular fuzzy formation of a family of information granules over given Analytical Hierarchy Process (AHP) numerical scales. On this basis, the individual consistency and group consensus measure indices using fuzzy granulation technique are constructed, respectively. Then, the optimal cut-off points of fuzzy information granules are obtained by establishing a multi-objective optimization model together with a multi-objective particle swarm optimization (MOPSO) algorithm. A novel group consensus decision-making approach where consensus reaching process (CRP) is achieved by adaptively adjusting individual preferences through the optimization of the cut-off points is proposed. After conflict elimination, the obtained group preference gives the ranking of the alternatives. Finally, a real emergency decision-making case for liquid ammonia leak is given to illustrate the application steps of the proposed method and comparative analysis with the existing GDM methods. Comparative results demonstrate that the proposed method has some advantages in aspects of avoiding information loss or distortion and improving consensus performance.     

Multirobot Object Localization: A Fuzzy Fusion Approach

In this paper, we address the problem of fusing information about object positions in multirobot systems. Our approach is novel in two main respects. First, it addresses the multirobot object localization problem using fuzzy logic. It uses fuzzy sets to represent uncertain position information and fuzzy intersection to fuse this information. The result of this fusion is a consensus among sources, as opposed to the compromise achieved by many other approaches. Second, our method fully propagates self-localization uncertainty to object-position estimates. We evaluate our method using systematic experiments, which describe an input-error landscape for the performance of our approach. This landscape characterizes how well our method performs when faced with various types and amounts of input errors.      

不确定性研讨信息的结构化与融合方法

基于流形正则化框架提出一种分类算法(MLD-RLSC),以解决高维文档分类问题。该算法通过构建训练样本的最近邻图来估计数据空间的几何结构并将其作为流形正则化项,结合多变量线性回归获得高维文档的低维流形结构,并采用k近邻分类器对低维流形进行分类,得到针对多类问题的分类器。该算法能够充分利用训练样本的类别信息来帮助学习以提取有效特征。通过在Reuters-21578数据集上的实验,证明该算法的分类性能和运行速度比传统分类器有较大的提高。     

基于共识与Choquet积分的模糊Petri网可信度确定方法*

针对模糊Petri网可信度确定过程中,决策结果可接受程度较低和未考虑影响因子间关联的问题,提出基于群体共识测度与Choquet积分确定模糊Petri网可信度的方法。设计了基于决策者与群体评估信息偏差修正的自适应共识测度算法,获取可接受程度较高的群体评估信息。构建最大2-可加模糊测度Marichal熵模型求解影响因子间的交互作用系数,根据交互作用系数、默比乌斯变换和模糊测度之间的关系确定模糊测度,利用Choquet积分算子集结群体评估信息得到模糊Petri网可信度。将此方法用于确定燃气轮机故障诊断模糊Petri网输入库所可信度,并与未考虑共识测度和影响因子间关联的方法进行比较,计算结果验证了该方法的有效性和可行性。     

一种基于相容关系的群体共识达成方法

针对一类评价信息不完全的多属性群决策问题中的群体判断的共识性进行研究,提出一种基于相容关系的群体共识达成方法。该方法不需进行群体共识度的计算,避免了评价信息不完全时直接计算群体共识度的困难;对于专家没有给出的评价信息,不需要进行评估和填充,避免了评估和填充缺失值时所需进行的复杂计算。最后通过一个实例对所提出方法的有效性和实用性进行了说明。     

Multi-attribute group decision making with opinion dynamics based on social trust network

Social trust network (STN) has facilitated information exchange between experts during interactions. Some feedback mechanisms have been used to provide advices for opinion change to improve their consensus levels. However, they do not consider the experts’ willingness and their self-confidence values. To analyze the influence of the relationship between experts on the decision-making results, this paper proposes a multi-attribute group decision making (MAGDM) with opinion dynamics based on STN. Three stages are included in the proposed approach: trust propagation, consensus reaching process and alternative selection. In the trust propagation stage, the social weight influence matrix and the weights of experts are obtained based on the complete social trust matrix which is constructed by trust aggregation and the given self-confidence values of experts. In the consensus reaching process, the consensus measure is used to determine the consensus between the experts or not, and the feedback mechanism based on opinion dynamics is used to adjust the opinions which do not reach consensus. The appropriate alternative is selected based on the assessable value of the alternative in the selection process. Finally, a numerical experiment about supplier selection is introduced to illustrate the efficiency of the proposed approach and comparison analyses show that the proposed approach can improve efficiency compared with the MAGDM in the social network.