改进的图像分割方法的应用

为了实现较好的在复杂背景下跟踪目标,提出一种基于YUV灰度阈值法和基于Gauss Laplacian边缘检测法相结合的算法。该算法首先采用灰度阈值法对图像进行灰度处理,使图像只有目标颜色和背景颜色2种颜色,再用Gauss Laplacian边缘检测提取目标轮廓,根据目标形状特征去除与目标颜色相同的其他物体。实验表明与以往采用单一方法相比,该方法在复杂背景下可以很好的提取出目标,达到准确跟踪的效果。     

不可约Z-矩阵最小特征值的数值算法

首先给出了不可约非负矩阵最大特征值的上下界。然后利用相似变换构造了一列相似矩阵,从而得到不可约非负矩阵最大特征值的逐步压缩的一列上下界,其极限为所要求的最大特征值。最后利用Z-矩阵与非负矩阵的关系,给出了计算不可约Z-矩阵最小特征值的一个新算法。理论上给出了收敛性证明。该算法迭代过程简单,不用计算逆矩阵,从而计算量小,占用内存少。数值实验的结果表明该算法具有可行性和有效性。     

基于Laplacian Eigenmap的图像变化检测虚警优化技术

对点目标的图像变化检测,现有的变化检测技术结果往往存在着虚警过大的问题。通过深入分析多个传统的变化检测方法的特点,利用各方法的互补性,提出了利用Laplacian Eigenmap对多个方法检测结果进行降维分类的优化技术。首先把各个方法对某个像素的检测结果用向量的形式进行表示,然后利用Laplacian Eigenmap对整个图像的数据流形在低维空间展开,最后利用模糊分类进行分类。该技术有两个优势：（1）在保证现有较高检测率的同时,大大降低了结果的虚警率;（2）它极大地降低了在传统方法中由于人为阈值取舍带来的偏差风险。但该技术的不足之处是增加了计算量。     

Application of Component Mode Synthesis to Protein Structure for Dynamic Analysis

This paper concerns the application of component mode synthesis for biomolecule modeling to understand protein dynamics. As for protein dynamics, eigenvalue problem should be formulated to obtain eigenvalue, eigenvector and thermal fluctuation. To describe the thermal fluctuation of protein, normal mode analysis is introduced and normal modes identify the dynamic behavior of protein very well. Component mode synthesis considers the given complex structure as an assembly of smaller components. The selection of a component may be arbitrary. When the component mode synthesis is applied to formulate the eigenvalue problem of protein structure, we selected a protein which may be composed of two and/or four domains. The domain of protein can be considered as a component. In this sense the number of component is increased as necessary, and the size of each component is decreased for fast calculation. The component mode synthesis widely used in engineering was well applied to understand protein dynamics in present study.     

水轮机及其调速系统对电力系统低频振荡的影响

传统上在对电力系统低频振荡的研究中,励磁系统被认为是造成低频振荡的主要原因。而目前认为,水轮机及其调速系统对电力系统低频振荡的影响也不能忽略。对低频振荡的研究主要从水轮机及其调速系统的作用着手,通过对实例的计算得出特征值、相关因子等,通过相关性分析等方法得出水轮机及其调速系统对电力系统低频振荡有影响的结论。     

基于群论的预应力结构广义特征值分析

由于常规自振分析方法未能充分利用结构的固有对称性,当结构自由度增加时,计算消耗显著提升。本文基于群论方法,提出一种分析对称型预应力结构动力特性的高效方法。首先,结合一致质量矩阵及切线刚度矩阵建立了预应力结构的广义特征方程,并考虑初始预应力对结构的影响,以求解结构的自振频率及振型。随后,建立对称坐标系,将刚度矩阵及质量矩阵分解为一系列分块对角化矩阵。由于各分块子矩阵相互独立,广义特征值问题的求解难度显著降低,从而能高效求解结构的自振频率及相应振型。数值算例阐明了所提出方法的基本计算过程及显著优势。与有限元结果及常规方法所得结果比较后可知,基于群论的对称方法准确、高效。     

一类在两个斑块内人口迁移的传染病模型的研究

建立了一类在两个斑块内人口迁移的传染病模型，得到了该模型的基本再生数R12和主特征值λ1，证明了若λ1〈0，则无病平衡点是全局渐近稳定的，即疾病在人口迁移条件下消失，若λ1〉0，则地方病平衡点存在，且是全局渐近稳定的，即疾病在人口迁移条件下持续存在．     

A generalised back-flow model to predict the tracer response in liquid—liquid stagewise contactors

A generalized transient back-flow model to predict the transient tracer concentration in a multistage extraction column has been developed. The model does not assume any extra stage at the end of the contactor and is based on unified formulation. The solution can be obtained in a straight forward way by using the linear operator theoretic approach as opposed to Laplace transform.     

Identification of characterization factor for power system oscillation based on multiple synchronized phasor measurements

Electric power systems in Japan are composed of remote and distributed location of generators and loads mainly concentrated in large‐demand areas. The structures having long‐distance transmission tend to produce heavy power flow with increasing electric power demand. In addition, some independent power producers (IPP) and power producer and suppliers (PPS) are participating in the power generation business, which makes power system dynamics more complex. However, there was little observation as a whole power system. In this paper the authors present a global monitoring system of power system dynamics by using the synchronized phasor measurement of demand‐side outlets. Phasor Measurement Units (PMU) are synchronized based on the global positioning system (GPS). The purpose of this paper is to show oscillation characteristics and methods for processing original data obtained from PMU after certain power system disturbances triggered by accidents. This analysis resulted in the observation of the lowest and the second lowest frequency mode. The derivation of eigenvalue with the two‐degree‐of‐freedom model brings a monitoring of two oscillation modes. Signal processing based on wavelet analysis and simulation studies to illustrate the obtained phenomena are presented in detail. © 2008 Wiley Periodicals, Inc. Electr Eng Jpn, 163(3): 10– 18, 2008; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/eej.20316     

An edge preserved image compression technique

An image compression technique that can preserve edges of objects is proposed here. This technique is developed by looking at small local regions of the image and classifying these regions as uniform, point, line or edge patterns from statistical likelihood ratio tests. The compression ratio depends on the size of the local region considered and the quality of the image will degrade as the size of the local region increases. Experimental results show good quality preservation when the sizes of the local region are 3 × 3, 4 × 4, 5 × 5 with the corresponding compression ratios less than or equal to 7/24, 1/5, 13/100 respectively. A comparison with discrete cosine transform is given here and shows that the new technique produces better quality in most cases.