Analysis of Two-Dimensional Non-Rigid Shapes

Analysis of deformable two-dimensional shapes is an important problem, encountered in numerous pattern recognition, computer vision and computer graphics applications. In this paper, we address three major problems in the analysis of non-rigid shapes: similarity, partial similarity, and correspondence. We present an axiomatic construction of similarity criteria for deformation-invariant shape comparison, based on intrinsic geometric properties of the shapes, and show that such criteria are related to the Gromov-Hausdorff distance. Next, we extend the problem of similarity computation to shapes which have similar parts but are dissimilar when considered as a whole, and present a construction of set-valued distances, based on the notion of Pareto optimality. Finally, we show that the correspondence between non-rigid shapes can be obtained as a byproduct of the non-rigid similarity problem. As a numerical framework, we use the generalized multidimensional scaling (GMDS) method, which is the numerical core of the three problems addressed in this paper.     

基于LSSVM的木材干燥建模研究

针对木材干燥过程的强非线性特点,提出以最小二乘支持向量机LSSVM建立木材干燥基准模型.通过实验用小型木材干燥窑实际干燥过程中采集的数据作为训练样本进行仿真实验,结果表明基于LSSVM的木材干燥模型预测输出能够准确反映干燥过程木材含水率的变化,模型结构简单、预测精度高、泛化能力强,验证了LSSVM对木材干燥过程建模是一种可行而有效的方法.     

MIMO-CDMA子选择系统设计及性能分析

多用户MIMO-CDMA是MIMO技术领域研究的最新成果,MIMO-CDMA系统可以显著地提高系统的容量和BER性能。但系统复杂度高,而天线选择技术可以大大降低MIMO-CDMA系统实现复杂度。设计了一种基于帧结构的MIMO-CDMA接收端子选择系统仿真平台,分析和比较了在MBER、MMI和MNP等不同选择准则条件下的系统性能,并对仿真结果进行了评价。     

GM(n,h)模型建模序列数据数乘变换特性研究

基于数乘变换是灰色系统建模过程中数据处理的基础,讨论了ＧＭ（n,h）模型与其他几类灰色模型的内在联系．将各类灰色模型统一于共同的分析体系,并在此基础上研究了数乘变换对ＧＭ（n,h）模型参数取值的影响．指出了模型的模拟和预测值只与因变量的数乘变换有关,而与自变量的变换无关．最后分析了几个特殊灰色模型的数乘变换性质,该结果对研究系列灰色模型参数特征有重要意义．     

一种改进的在线最小二乘支持向量机回归算法

针对一般最小二乘支持向量机处理大规模数据集会出现训练速度幔、计算量大、不易在线训练的缺点,将修正后的遗忘因子矩形窗方法与支持向量机相结合,提出一种基于改进的遗忘因子矩形窗算法的在线最小二乘支持向量机回归算法,既突出了当前窗口数据的作用,又考虑了历史数据的影响.所提出的算法可减少计算量,提高在线辨识精度.仿真算例表明了该方法的有效性.     

Robust supervisory control of timed discrete event systems under partial observation based on eligible time bounds: The existence conditions

This paper addresses a supervisory control problem for uncertain timed discrete event systems (DESs) under partial observation. An uncertain timed DES to be controlled is represented by a set of possible timed models based on the framework of Brandin and Wonham [(1994). Supervisory control of timed discrete event systems. IEEE Transactions on Automatic Control, 39(2), 329-342]. To avoid the state space explosion problem caused by tick events in the timed models, a notion of eligible time bounds is proposed for a single timed model obtained from the set of all possible timed models. Based on this notion, we present the necessary and sufficient conditions for the existence of a robust supervisor achieving a given language specification for the single timed model. Moreover, we show that the robust supervisor can also achieve the specification for any timed model in the set.     

Supervisory control of modular systems with global specification languages

The paper presents sufficient conditions for modular (supervisory) control synthesis to equal global control synthesis. In modular control synthesis a supervisory control is synthesized for each module separately and the supervisory control consists of the parallel composition of the modular supervisory controls. The general case of the specification that is indecomposable and not necessarily contained in the plant language, which is often the case in practice, is considered. The usual assumption that all shared events are controllable is relaxed by introducing two new structural conditions relying on the global mutual controllability condition. The novel concept used as a sufficient structural condition is strong global mutual controllability. The main result uses a weaker condition called global mutual controllability together with local consistency of the specification. An example illustrates the approach.     

泛函网络神经元构造理论与方法

泛函网络是近年提出的一种对神经网络的有效推广.与神经网络不同,它处理的是一般的泛函模型,它在各个神经元之间的连接没有权值,并且神经元函数不固定的,往往是一给定的基函数的组合,泛函网络学习的目的就是求出神经元函数的精确表达式或近似表达式. 迄今关于泛函网络神经元基函数的存在性和选取方法缺乏理论依据.文中基于Banach空间中偏序理论,分析了泛函网络神经元基函数的存在性,给出了泛函网络神经元基函数选取方法,对于完善泛函网络的基础理论具有参考价值.     

一阶双曲构建Hill密码及计算机模拟

讨论用PDEs构建Hill密码的方法。以一阶线性非齐次双曲方程混合问题的形式给出加、解密问题的模型,由差分格式算法设计可用于加、解密的矩阵方程。改进的Hill密码系统中,矩阵变化多样、密钥空间大且便于传输和管理。用MatLab编制软件实现加、解过程并对部分结果进行分析。     

基于K0honen网络和OLS算法的RBFNN中心选择方法

提出了一种优化选择径向基神经网络数据中心的算法,该算法结合了Kohonen网络的模式分类能力,将初步分类结果用作RBFNN的初始数据中心,然后采用OLS算法进行优化,对比仿真实验表明该算法效果比单独使用OLS算法生成的RBFNN性能更好。