基于模糊隶属度的人脸识别应用*

针对人脸图像特征提取,应用主成分分析和二维主成分分析方法,提出用二维特征求解样本的隶属度,用主成分特征进行支持向量机分类的方法。该方法结合了二维主成分特征在选取少量分量时人脸重构图像稳定的优点和主成分特征重构图像局部特征清晰的优点。为了与二维主成分特征分类结果进行比较,通过引入矩阵内积,给出了针对二维特征的三类核函数。实验表明利用两种特征进行分类的方法在人脸识别中具有较高的精度。     

Characterisation of ecotones using membership degrees computed with a fuzzy classifier

Ecotones zones lie between homogeneous ecological systems. They are characterised on images by heterogeneous pixels in a specific neighbourhood. Fuzzy classifiers output membership degrees that better represent the heterogeneity within a pixel and can be further processed within the context of a local neighbourhood. This Letter formalises these notations. Two coral reef systems examples are presented. They illustrate the use of possibility measurement to characterise ecotones, and the use of information on well-known ecotones to increase the accuracy of image classification.     

A relationship between membership functions and approximationaccuracy in fuzzy systems

This paper presents a relationship between membership functions and approximation accuracy in fuzzy systems. This relationship suggests an idea to design membership functions such that the approximation accuracy of fuzzy systems is improved.     

不确定隶属函数T-S模糊控制器设计与稳定分析

针对许多非线性系统存在的结构不确定和难以用精确数学模型表达等问题,在研究基本Takagi-Sugeno(T-S)模糊模型的基础上,通过增加一个隶属函数自由变量,利用线性矩阵不等式(LMI)设计了一个使得闭环系统渐进稳定的状态反馈控制器,并给出稳定条件.所得到的条件充分利用了前件变量隶属函数的结构信息和后件局部子系统之间的相互关系,降低了常规T-S模糊系统的稳定性条件的保守性和求解难度.通过仿真实例验证了该方法的可行性和有效性.     

基于混合隶属度的模糊简约双支持向量机研究

双支持向量机是一种新的非平行二分类算法,其处理速度比传统支持向量机快很多,但是双支持向量机在训练之前要进行大量的复杂逆矩阵计算;在非线性情况下,它不能像传统支持向量机那样把核技巧直接运用到对偶优化问题中;并且双支持向量机没有考虑不同输入样本点会对最优分类超平面产生不同的影响。针对这些情况,提出了一种模糊简约双支持向量机。该模糊简约双支持向量机通过对二次规划函数和拉格朗日函数的改进,省略大量的逆矩阵计算,同时核技巧能直接运用到非线性分类情况下;对于混合模糊隶属度函数,不仅每个样本点到类中心的距离影响着该混合模糊隶属度,而且该样本点的邻域密度同样影响着该混合模糊隶属度。实验结果表明,与支持向量机、标准双支持向量机、双边界支持向量机、模糊双支持向量机相比,具有该混合模糊隶属度函数的简约双支持向量机不仅分类时间短,计算简单,而且分类精度高。     

基于置信隶属度的多传感器数据模糊融合

在数据的融合过程中,将每个传感器的数据看成服从正态分布,利用置信水平得到隶属度,构造优越性比较矩阵,进而得到融合后的数据,将该方法运用于实证,并与平均权、概率权等方法进行比较,实证部分验证了该方法的实用性,再进行多次仿真,通过绝对误差的比较,验证了此方法的稳定性和有效性.     

基于模糊隶属度的支持向量机去噪方法

针对传统支持向量机对于噪声和野点敏感的问题,采用一种模糊技术去除样本中的噪声和野点.应用基于样本之间的紧密度确定每个样本的模糊隶属度,通过训练确定阀值,去除影响得到最优分类超平面的噪声和野点.实验结果表明,与传统的支持向量机相比,该方法提高了支持向量机的抗噪能力,在不影响精度的前提下,线性规划下的一类分类方法要比二次规划节省很多时间.     

基于加权模糊隶属度的二叉树多分类算法

针对二叉树支持向量机多分类算法准确率与分类效率较低的问题,提出了一种基于加权模糊隶属度的二叉树支持向量机多分类算法（binary tree support vector machines multi-classification algorithm based on weighted fuzzy membership,PF-BTSVM）。该算法依据最大最小样本距离与质心距离构造出一个近似完全二叉树,提高了整体结构的分类效率;利用模糊隶属度函数以及正负辅助惩罚因子对训练集进行筛选,剔除掉对分类无用的样本与噪声值,实现了训练集的提纯并且削弱了不平衡分类时超平面的偏移。在数据集上的实验结果表明,与其他二叉树多分类算法相比,该算法在提高了分类准确率以及稳定性的的同时还加快了训练与分类的速度,而且这种优势当分类的不平衡度越大时越明显。     

基于模糊聚类的车辆跟驰隶属度函数确定方法

当前研究确定车辆跟驰模糊推理隶属度函数时所采用的方法主要是专家法,不能精确获得车辆跟驰隶属度函数。针对于此,提出根据模糊聚类分析的方法,考虑车辆跟驰数据内部的关联性,利用基于高斯函数的隶属度函数确定方法,进行车辆跟驰模糊集的划分和隶属度函数的确定。使用真实的车辆轨迹数据,将后车速度、前后车相对速度、车间距作为输入变量,后车加速度作为输出变量建立模糊推理系统,对论文提出的基于模糊聚类的车辆跟驰隶属度函数确定方法进行评价。结果表明：本文提出的新方法能真实反映数据本身的特征和驾驶员的心理生理特性,其推理结果与真实数据误差较小,可用于分析模糊推理的车辆跟驰行为特点。     

基于可分性测度的模糊隶属函数确定方法

隶属程度的思想是模糊数学的基本思想,应用模糊数学方法解决实际问题的关键在于建立符合实际的隶属函数,然而,如何正确地确定隶属函数仍是至今尚未完全解决的问题.鉴于此,提出一种基于可分性测度的隶属函数确定方法,利用类间在各个特征上的可分性确定模糊集的划分,进而确定描述该模糊集的隶属函数.通过轨道电路故障诊断实验表明了所提出方法的有效性.     

基于最小二乘拟合的模糊隶属函数构建方法

针对当前模糊隶属函数构造方法中存在的问题,提出一种构造模糊隶属函数方法.采用最小二乘法拟合离散数据来获得隶属函数.为减小拟合误差,采用了3项措施以达到预期目标.所构建的隶属函数,对任意输入物理量可直接得到其对应模糊语言变量的隶属度,从而有效避免专家指定隶属度的主观臆断性及不一致性.该方法简单、求解精度高,具有广泛适用性和较强的应用价值.仿真结果证实了该方法的有效性.     

基于遗传算法的模糊隶属函数的选择

确定合适的隶属函数对模糊控制效果有重要影响,是应用模糊控制的基础,也是难点之一。作者应用遗传算法对模糊温度传感器信号处理过程中的隶属函数进行选择,阐述了基本原理和具体实施过程,最后用一具体的实例来说明该优化算法在模糊隶属函数选择中的应用。结果表明：该方法简单实用,具有较为理想的效果。     

A multi-criteria route planning model based on fuzzy preference degrees of stops

Integrated utilization of new technologies such as smart phones, tablet devices, and satellite maps has entered our daily lives recently. Nevertheless, many new applications are being developed mostly based on these technologies. The optimal route planning, which makes use of the public transport network structure between any selected origin and destination points, is one of the interesting applications among them. Route planning applications used today mostly focus on the aspects such that passengers use nearest stops around origin and destination geographical points, or use set of stops around these points within some walking radius. In these applications, which work on the classical (crisp) logic base, all stops on the walking distance have the same preference degree. However, in this study a novel fuzzy model is proposed which also takes into account preferences such as the stop’s activity, and count of transit lines passing through the stop besides the walking distance. Using all these three preferences, aggregated fuzzy preference degrees of stops are calculated. The “optimum” routes between any origin and destination pair are constructed using feasible transfer points, which are chosen among the alternatives having the highest preference degrees overall. Fuzzy neighborhood relations such as “stop-stop”, “stop-line”, and “line-line” are introduced in order to employ in preference degree evaluations.Apart from the aggregated degree of the preferences mentioned above, we also consider to minimize the total number of transit stops travelled on any route for establishing optimal routes. This additional preference can be described the time duration spent on transport vehicles, such as buses, trains, subways or ferries. Therefore, we propose a two-criteria route-planning problem in this study, where we try to maximize the aggregated preference degree of a route and to minimize the number of stops used on a route. Fuzzy optimal solutions for this problem are constructed via γ-level solutions of the fuzzy problem and a heuristic algorithm providing these solutions is proposed. This model and its algorithm can be considered as an optimal route search engine for mobile applications that could be used by urban public transport passengers.     

基于小波隶属函数的模糊推理规则优化

隶属函数决定着模糊集的特征，建立小波基函数与隶属函数之间的联系，从而利用小波分析探讨模糊推理的实质，以一种非对称Haar小波基与三角型、梯型隶属函数的对应关系为基础，将小波分析、遗传算法与模糊系统结合，利用遗传算法实现小波隶属函数的训练学习，进而实现模糊推理规则的优化。     

一种基于航迹隶属度的动态加权融合算法

提出一种适用于传感器精度差异较大时的目标精确跟踪的加权融合算法,该算法通过求解局部航迹之间的最佳隶属度矩阵,将局部航迹本身的最佳隶属度动态地赋给对应传感器权值,实现权值的动态分配。蒙特卡洛仿真表明,精度差异较大传感器组成的融合跟踪网络中,该算法不论应用于匀速目标还是机动目标的精确跟踪,均具有良好的性能。     

基于小波隶属函数的模糊推理规则优化

隶属函数决定着模糊集的特征,建立小波基函数与隶属函数之间的联系,从而利用小波分析探讨模糊推理的实质,以一种非对称Haar小波基与三角型、梯型隶属函数的对应关系为基础,将小波分析、遗传算法与模糊系统结合,利用遗传算法实现小波隶属函数的训练学习,进而实现模糊推理规则的优化。     

基于包含度的直觉模糊相似度量方法

针对现有文献中涉及到直觉模糊相似度量的计算公式大多是基于直觉模糊距离测度的现状,提出一种基于包含度的直觉模糊相似度量方法。借助模糊蕴涵算子和集合基数,建立了IFS包含度的一系列具体公式,揭示了IFS包含度与相似度的关系,提出了一种基于包含度的IFS相似度量方法,并以具体算例表明该方法不仅可以解决中部分存在的问题,而且为研究IFS相似度量理论开辟一个新的途径。     

A method for automatic generation of fuzzy membership functions for mobile device’s characteristics based on Google Trends

While creating a framework for adaptive mobile interfaces for m-learning applications we found that in order to ease the use of our framework we needed to present the mobile device characteristics to non-expert users in a easy to understand manner. Using fuzzy sets to represent the characteristics of mobile devices, non-expert developers such as teachers or instructional designers can actively participate in the development or adaptation of the educational tools. To be able to automatically generate the fuzzy membership functions of the sets we needed the data of the mobile device market, regrettably this information is not publicly available. To tackle this problem we have developed a method to estimate the market share of each mobile device based on the popularity metrics recovered from Google Trends and then we use that estimated value as the input to generate the fuzzy set of each characteristic. The proposed method allows us to not only model the state of the market in different periods of time, but also to localize the results to adapt them to the mobile market of specific countries. In this paper we will describe the proposed algorithm and we will discuss the obtained results.     

基于局部隶属度和邻域信息的GIFP-FCM图像分割算法

基于一般化的模糊划分GIFP-FCM聚类算法是模糊C均值算法(FCM)的一种改进算法,一定程度上克服了FCM算法对噪声的敏感性,但由于其没有考虑图像的邻域信息,对含有较大噪声的图像分割效果不理想。为此,提出将局部隶属度和局部邻域信息等引入到GIFP-FCM算法的目标函数中,通过重新计算每个像素的局部隶属度和邻域信息,较好地克服了噪声影响。利用该算法对合成图像、脑图分割的实验结果表明,对于含有高斯噪声、椒盐噪声和混合噪声的图像,新算法得到的划分系数值最大,划分熵最小,是一种去噪效果较好的图像分割算法。