独立分量分析的图像融合算法

独立分量分析可实现图像的稀疏编码并具有能很好地捕捉图像重要边缘信息的特性.本文提出一种基于独立分量分析的图像融合算法,结合支持向量机对多聚焦图像的清晰域、模糊域进行判断以及在ICA域中进行图像分割以提取图像的主要边缘特征信息来实现特征级的多聚焦图像的融合.实验结果表明,本文提出的融合算法是有效的.     

基于独立分量分析的荧光造影图像降噪算法

荧光造影图像分析是一种重要的糖尿病眼底病变测试方法,针对荧光造影图像易受噪声污染以及白内障病变影响导致眼底视网膜图像辨识困难等问题,基于盲信号处理理论,运用维纳滤波与独立分量分析进行降噪,并运用FASTICA算法进行源信号提取.测试图像处理结果表明:该方法能够有效地抑制噪声,提高分离图像的信噪比,且较好地恢复原始图像.     

A parallel mesh generation algorithm based on the vertex label assignment scheme

In this article a new mesh generation algorithm is presented. The algorithm is based on a new approach called the vertex label assignment scheme to provide the information for the mesh generation so that parallel processing becomes possible. The algorithm generates 2D meshes of quadrilaterals on the basis of individual faces; conformity and smoothness of the resultant mesh are automatically assured. Local and selective mesh-refinements are also supported. A regular quadrilateral network which defines the geometry of the problem and an associated subdivision level assignment which specifies mesh density data on the network are the only input information.     

基元树建筑物图像伪造组件检测算法

为解决当前图像伪造检测方法因采用逐像素或逐块比对的鉴别方式而导致耗时长、内存空间消耗大以及伪造定位准确度较低等不足,提出基于基元树的建筑物图像伪造组件检测算法。该算法根据建筑物的对称性特征,提取目标图像的基元,获取包络盒图像,再以其为根结点,借助分割与嵌入技术,对组件进行拆分,构造出基元树;再利用基元树来表征建筑物图像,凸显基元之间的空间分布关系,并定义非聚合组件的对称因子差异以及聚合组件的密集度差异计算模型,通过检索基元树结点对应的非聚合组件对称因子差异值或聚合组件密集度差异值,快速检测出建筑物图像中的疑似伪造组件,即疑似伪造图像区域。实验结果表明:与当前图像伪造检测技术相比,所提方法的检测速度更快,约19.92s;且准确度更高。所提基于基元树的建筑物图像伪造组件检测算法能够高效准确检测出建筑物图像伪造区域。     

基于申威异构众核处理器架构的模态并行算法

根据国产申威异构众核处理器架构特点,提出了一种结构有限元模态分层通信并行计算方法,对于提高国产申威异构众核分布式存储并行计算机下重大装备系统级模态分析的并行效率具有重要意义.该方法在分层通信策略和加速子空间迭代法的基础上构建了大规模模态分析并行计算体系,不仅实现了计算过程和数据通信的分层,有效提高了通信效率;而且实现了...     

Development of a collaborative product development framework based on centre-satellite system and service-oriented architecture

To increase competitive advantage for manufacturers, collaboration product development (CPD) has recently become an important issue. Based on the CPD concept, this study focuses on the centre-satellite system (CSS) with five major roles (i.e. centre factory) and their interactions and the technology of web services using service-oriented architecture (SOA) to propose a system framework, entitled the collaborative product development framework for centre satellite system (CPDF-CSS). The CPDF-CSS comprises five modules: CF-MA module, DH-MA module, ODM-MA module, OEM-MA module, and EC-MA module. Next, based on the CPDF-CSS, a generic communication component (GCC) is designed for the interoperability among five modules under heterogeneous contexts. Finally, this study presents a case involving the motorcycle industry for evaluating the implementation of CPDF-CSS. The contributions of this study to the CPD are discussed.     

新一代互联网服务模型及框架研究

针对下一代互联网基础网络资源异构、用户需求个性化、参与者众多等特征,结合Web服务、语义Web、服务分类、参与者角色细分、面向服务的体系结构、开放网格服务体系结构等研究成果,分别从服务参与者、语义描述、技术视点角度提出了多角色合作交互管理模型、本体模型和能力分层提供模型.多角色合作交互管理模型使用智能化方式综合管理服务,本体模型解决服务本质特征统一语义描述的问题,能力分层提供模型实现异构基础网络资源的融合开放,并基于这三个服务模型提出了一个完整的服务框架.     

基于虚拟化的轻量级背景流量节点生成及部署算法研究

进行了网络攻防分析,针对传统的背景流量模拟技术在生成及部署方面存在背景流量大小}限、背景流量真实性不足、背景流量生成工具部署的数量和位置}限的问题,提出了一种基于虚拟化的轻量级背景流量节点生成及部署算法,该算法是一种基于资源量的目标流量映射、基于最短路由的背景流量应用添加、基于最少通信代价的节点映射的三阶段算法,解决了背景流量节点在网络攻防实验中大规模快速生成及灵活部署的问题。实验结果表明,使用轻量级背景流量节点,生成的背景流量满足网络自相似性,资源消耗小、启动速度快,并且可根据实际网络攻防实验的需求进行动态、灵活、快速的大规模部署。     

改进的基于独立成分分析的图像特征提取算法

利用基函数稀疏性的极大值特点,本文提出一种改进的基于独立分量分析的图像特征提取算法.通过分析图像的拉普拉斯先验条件,独立成分分析的问题简化为L1范数求最小的问题,但是采用其对偶空间L∞范数求极大值的方法解决问题更加容易.相对其它独立成分分析方法,本文方法不需要对高阶非线性对比函数估值进行复杂的优化.实验结果表明,本文算法相比其它独立成分分析算法,具有更好的稀疏性和更快的收敛速度.     

A morphing method for shape generation and image prediction in product design

A shape morphing and image prediction method for product design is proposed in this article, in which a feature-based method is first used to construct 3-D CAD models of a product, then new shapes are generated with a shape morphing method and the images of the morphed shapes are analyzed using modified gray theory with Fourier residual correction. With the aid of this model, the designer can quickly obtain a product form and its corresponding image to help him/her to design the required product. Though LCD monitor design is applied, this method can also be used to develop other products.     

基于自适应QPSO算法的软件测试数据自动生成

针对软件测试数据采用遗传算法和粒子群算法自动生成算法复杂和容易早熟等问题,提出一种动态调整收缩扩张因子的自适应量子粒子群算法(AQPSO)。该算法通过引入粒子进化度和聚合度,收缩扩张因子随粒子进化度因子和聚合度因子变化而变化,从而实现算法的动态自适应性,提高算法收敛速度和求解精度。软件测试数据自动生成实验验证了该算法的有效性和可行性。     

A new instrument and technique for diagnosing electromagneticdesign problems in microwave module housings and component packages

A method of measuring the frequencies at which resonances occur and the distribution of the associated electric field intensity at the inside face of the housing or package lid has been developed and tested. The method does not require modification of the connections to the microwave assembly within the housing. It is only necessary that the housing or package have a flat lid and that it be left off so that the item under test can be mounted on an equivalent flat plate on the test instrument. The microwave assembly in the housing or package may be operated normally while resonant field distributions are being measured. The prototype instrument yields results that demonstrate the practical application of this measurement assembly as a diagnostic and design development tool for microwave housings and packages of rectangular metal box basic shape. Resonances in the 8-12 GHz range have been accurately measured and indicate that by scaling dimensions an assembly for use at millimeter wavelengths is feasible     

A corner preserving algorithm for realistic DEM soil particle generation

An efficient particle clump generation algorithm for use in discrete element methods (DEMs) was developed to simulate actual particles in a granular specimen. The algorithm requires many fewer circles than existing methods, particularly for angular particles. The procedure is a logical extension of a computational method for determining the classic Wadell particle roundness in which circles are fitted to the corners of particles. Hence, the new method perfectly preserves the location, size and shape of particle corners and is appropriately termed the corner preserving algorithm. The remaining perimeter of the particle, which includes concave and flat sections, is fitted with non-corner circles. Unlike earlier methods, the corner preserving algorithm requires only a single control parameter to generate the circles. This parameter is the ratio of the clump area to the original particle area, AR. An algorithm seeks a corresponding optimum clump roughness that achieves the user-prescribed AR. The method can easily be incorporated in existing soil particle characterization systems in which binary images or even images of particle assemblies are produced. Examples illustrate the simplicity and advantages of the corner preserving algorithm for DEM clump generation.     

A hybrid model of component sharing and platform modularity for optimal product family design

Today's industry faces new challenges such as diverse customer demands, shorter product development cycles and cost pressure, which compel manufacturing firms to change their production paradigm from one-size-fits-all mass production toward mass customisation. Over the past decades, modular design has received great attention as a key enabler for mass customisation, and component sharing and platform modularity have been quite popular strategies for modular design. While modular design approaches and their strategies offer a number of advantages such as late product differentiation and changeability, there are unfortunately negative aspects, for example, sales loss due to reduced performance compared to integral design approaches, which have received little attention. Therefore, we propose a hybrid model of the two strategies in order to develop the most profitable product family. A detailed numerical analysis provides empirical support for the feasibility and effectiveness of the hybrid model.     

Integrated planning for product module selection and assembly line design/reconfiguration

To take full advantage of product modularity, modular product design and assembly system design/sreconfiguration have to be simultaneously addressed. The emerging reconfigurable production and flexible assembly techniques have made such an integrated approach possible. As such, this paper proposes an integrated approach to product module selection and assembly line design/reconfiguration problems. It further suggests that quality loss functions be used in a generic sense to quantify non-comparable and possibly conflicting performance criteria involved in the integrated problem. The complexity of the problem precludes the use of commercial software for solving meaningful sized problems in polynomial time. A genetic algorithm is therefore developed to provide quick solutions. An example problem is solved to illustrate the application of the proposed approach. Based on 72 randomly generated test problems, ANOVA analysis is further carried out to investigate the effects of genetic algorithm parameters. The convergence behaviour of the search processes is also examined by solving large problems with different numbers of operations and product modules.     

一种基于二阶统计量的盲信道辨识与均衡

本文提出了一种新的基于二阶统计量的盲信道辨识与均衡方法。通过对接收信号过采样,等效的信道矩阵有特殊的结构可以提供足够的信道信息给我们来估计信道矩阵。根据Tong等人的观点,本文给出了另外一种证明的方法,并由此得到了一种新的基于二阶统计量的盲信道辨识与均衡方法。仿真结果表明此方法和Tong的方法具有同样的均衡效果。     

服务组合中信任感知的成员服务选择算法

考虑到把信任度作为组合服务实现成员服务选择的依据可使组合服务的执行、调度更好地围绕服务的信任度展开和有利于信任度高的成员服务加入到服务组合的过程中来,进而提高组合服务的可信性,把信任引入到服务组合中,研究了服务组合中信任感知的成员服务选择问题,建立了选择问题的数学模型,并将选择问题归结为一个最小代价集合覆盖问题,同时建立了选择问题的有向图表示方法,并设计了基于蚁群系统的选择算法。实验结果分析表明,设计的信任感知的成员服务选择算法具有较好的寻优性能。     

Sweepline algorithm for unstructured-grid generation on two-dimensional non-convex domains

The sweepline algorithm has been modified to triangulate a set of nodes on a non-convex domain. The new method is based on a fragile sweepline which is split and rejoined in a manner which ensures that boundaries are preserved. Both the original and the modified algorithms are introduced using simple examples. It is shown that a set of N nodes is triangulated in O(N log N) time. The validity of the new method has been confirmed through testing on a variety of problems, and two test cases are presented in this paper.