ILUCP: a Crout ILU preconditioner with pivoting

We introduce a new preconditioner, ILUCP, to be used with an iterative method for solving sparse linear systems. It is based on an incomplete LU factorization combining Crout's formulation of Gaussian elimination with pivoting by columns. It is usually faster than ILUTP, which is based on a delayed update version of Gaussian elimination with pivoting, but requires more memory. For applications where memory is not a primary concern, ILUCP can be an attractive alternative to ILUTP. Copyright © 2005 John Wiley & Sons, Ltd.     

A multilevel Crout ILU preconditioner with pivoting and row permutation

In this paper, we present a new incomplete LU factorization using pivoting by columns and row permutation. Pivoting by columns helps to avoid small pivots and row permutation is used to promote sparsity. This factorization is used in a multilevel framework as a preconditioner for iterative methods for solving sparse linear systems. In most multilevel incomplete ILU factorization preconditioners, preprocessing (scaling and permutation of rows and columns of the coefficient matrix) results in further improvements. Numerical results illustrate that these preconditioners are suitable for a wide variety of applications. Copyright © 2007 John Wiley & Sons, Ltd.     

Robust Numerical Features for Description and Classification of Subcellular Location Patterns in Fluorescence Microscope Images

The ongoing biotechnology revolution promises a complete understanding of the mechanisms by which cells and tissues carry out their functions. Central to that goal is the determination of the function of each protein that is present in a given cell type, and determining a protein's location within cells is critical to understanding its function. As large amounts of data become available from genome-wide determination of protein subcellular location, automated approaches to categorizing and comparing location patterns are urgently needed. Since subcellular location is most often determined using fluorescence microscopy, we have developed automated systems for interpreting the resulting images. We report here improved numeric features for describing such images that are fairly robust to image intensity binning and spatial resolution. We validate these features by using them to train neural networks that accurately recognize all major subcellular patterns with an accuracy higher than any previously reported. Having validated the features by using them for classification, we also demonstrate using them to create Subcellular Location Trees that group similar proteins and provide a systematic framework for describing subcellular location.     

Design Method Based on Routing Tree for Topology Update in Ad Hoc Network

Ad hoc network is a kind of infrastructureless and self-organized mobile network. The wireless communication range of mobile nodes is limited in ad hoc network. The nodes can’t communicate with each other directly, but resort to the other adjacent nodes to forward their packets and exchange information of nodes. Every node not only serves as the mobile terminal, but also is able to store and forward packets[1]. Therefore, the node can be regarded as a router in ad hoc network. It finds the p…     

基于人工神经网络的通信信号分类识别

通信信号的分类识别是一种典型的统计模式识别问题。系统地论述了通信信号特征选择、特征提取和分类识别的原理和方法。设计了人工神经网络分类器,包括神经网络模型的选择、分类器的输入输出表示、神经网络拓扑结构和训练算法,并提出了分层结构的神经网络分类器。     

激光远场激发表面波在开口缺陷处的散射回波

为了实现激光超声技术对表面缺陷的定量检测,采用有限方法模拟了激光激发的表面波与表面开口矩形缺陷作用的复杂过程,比较了近场和远场两种不同激发方式下相同接收点的位移信号,发现远场激发下各种模式转换波和反射表面波能充分分离,有利于分析表面波与缺陷作用的复杂过程。在远场激发方式下分别探讨了表面波与缺陷前沿和后沿的作用过程,提取缺陷产生的散射回波特征,并用惠更斯原理分析其产生原因。最后研究了缺陷深度和宽度变化对散射回波特征的影响。结果表明,散射回波特征点到达时间差与缺陷深度和宽度有线性关系。该研究为反问题估算缺陷尺寸提供了理论依据。     

基于同一性的健壮CS分类算法

针对利用压缩感知(CS)进行信号分类识别的问题,提出了一种联合欲分类信号和样本信号的健壮CS分类算法。该方法通过引入“同一性”的概念,克服了信号过完备字典传统构造方式的不足,增强了信号稀疏表示与信号类别间的关联性,提升了基于压缩感知的信号分类算法性能。仿真实验证明了所提方法的正确性,并进一步表明:在非最优过完备字典下,该方法较之传统CS分类算法更具有分类准确度。     

基于多特征匹配的双向运动估计帧率提升算法

视频帧纹理区域含有的大量周期重复模式常导致双向运动估计发生视频块的错匹配,从而导致帧率提升算法重建的内插视频质量衰退.为了克服该问题,提出提取视频帧的多个特征加入到块匹配中,以降低错匹配发生的可能性.由于视频天然具有彩色信息,因此色差分量首先被加入到块匹配中.另外,人眼时图像边缘较敏感,因此,由简单的Sobel算子计算出的梯度被使用去反应边缘特征并融入至块匹配.多特征匹配尽管可有效地提高运动估计精度,但也引入了较高的计算复杂度.为了减少计算复杂度,一个特殊模板被设计去将多特征合并到单平面上,那么,只需进行一次块匹配运算就可完成多特征的匹配,在节省计算复杂度的同时也提高了运动估计精度.仿真实验表明,该算法以较低的计算复杂度获得了良好的内插帧主客观质量.     

基于自适应多字典学习的单幅图像超分辨率算法

自适应字典学习利用图像结构自相似性,将图像自身作为训练样本,通过字典学习使图像中的相似块在字典下具有稀疏表示形式.本文将全局字典学习中利用图像库获取附加信息的思想融入到自适应字典学习的过程中,提出了一种基于自适应多字典学习的单幅图像超分辨率算法,从低分辨率图像自身与图像库同时获取附加信息.该算法对低分辨率图像金字塔结构中的图像块进行聚类,在聚类结果的引导下将图像库中的图像块进行分类,利用各类中的样本分别构建针对各类的多个字典,从而确定表达重建图像块的最优字典.实验表明,与ScSR、SISR、NLIBP、CSSS以及mSSIM等算法相比,本文算法具有更好的超分重建效果.     

几何性特征的眉毛模型的研究

提出一种新的基于几何性特征的眉毛模型,相比传统的基于眉毛的全局数学特征建立的眉毛模型,它可以直观地描述和反映眉毛的几何外观形状,这种特性在人脸识别等基于面部视觉特征的应用是关键的.首先获取人脸单边眉毛图像并统一归一化尺寸为80×30的大小,使用阈值化和Blob分析得到眉毛归一化的二值图像,然后由二值图像计算眉毛模型的几何参数,包括眉毛区域的面积、长度、平均宽度、弯曲程度和走向趋势等.实验表明,提出的眉毛模型可以准确有效地描述眉毛的几何外观形状.