小波神经网络在图像配准中的应用

提出了一种基于小波网络的图像配准方法。将特征点定义进行了推广,提出了一种以特征区域定义和提取方法。使用Zernike矩表征区域的特征并进行特征区域的对应。因图像配准变换是复杂且难以预知的,利用小波神经网络具有良好的函数逼近性能,提出了具有局域特性的小波神经网络模型逼近图像的配准变换。实验表明这是一种有效的图像配准方法。     

一种基于区域分级合并的彩色图像分割方法

在区域合并过程中,手工设置颜色相似性和边界距离的权重极大地影响了分割的精度和自动化.针对这一问题,提出了一种新的基于区域分级合并的彩色图像分割算法.该方法能够根据邻接区域的边界特点设置权重因子,从而自适应地融合区域的颜色相似性和边界距离.使用均值漂移算法对图像进行初始分割,将原图像分割为具有较好边界的同质区域;通过计算区域相似度对区域进行分级合并.多幅彩色图像的分割实验结果证明,所提算法优于传统的基于区域合并的方法.     

Shear viscosity of the binary liquid mixture 3-methylpentane + nitroethane near the consolute point

Shear viscosity measurements of a mixture of 3-methylpentane and nitroethane at the critical concentration are presented in the temperature range 0.005 KT–T _c13.93 K, where T _c is the consolute or critical temperature. The data agree with earlier measurements.     

基于高斯映射下自由曲面的形状分析及边界计算

测定单一类型（凸或马鞍）区域和计算其边界是进行自由曲面的形状分析与控制研究的基本内容之一．目前这方面的研究局限于一些特殊曲面，如可展曲面，对自由曲面则无来境的理论和方法．本文提出了一些对复杂自由曲面进行形状分析的几何结论．通过计算高斯抛物线，划分了曲面的单类型区域．设计了算法判断单类型区域、分离单类型区域，并计算其边界，对自由曲面整体的形状有了较好的把握．最后蛤出了算法对两个例子的计算结果．     

基于可行域的遗传约简算法

在已有的遗传属性约简算法的基础上，通过引入约简的可行域概念，提出了基于可行域的遗传约简算法．可行域保持系统的分类能力，缩小了原问题的搜索空间，进而减小了问题的复杂度．适应度函数中引入与互信息相关的惩罚因子保证了算法在可行域中搜索．实验结果表明谊算法既克服了启发性算法的缺陷，较之已有的基于遗传算法的约简算法也有效率改进．     

基于遗传算法的发动机需求量组合预测方法

阐述了基于改进遗传算法组合预测的原理，给出了此类组合预测的建模方法及有效性评价指标。在军用航空发动机采购需求量的预测中，运用计算机仿真的方法对比了多种预测方法，验证了组合预测法的有效性和准确性。     

Sm2O3掺杂磷酸盐玻璃的形成及热稳定性

借助DSC分析手段,系统研究了稀土掺杂磷酸盐玻璃的形成及其影响因素,结果表明,当玻璃结构中[PO4]/[AlO4]≥2时易于形成均质透明的玻璃,形成区域范围为P2O5含量50%～85%(摩尔分数),BaO含量10%～40%(摩尔分数),Al2O3含量0%～17.5%(摩尔分数),Sm2O3含量0%～20%(摩尔分数);Al2O3含量增加,玻璃的形成能力增强,Sm2O3含量的增加使得玻璃的形成能力先增大后减小.     

Calibration of the Weibull stress scale parameter, σu, using the Master Curve

This work proposes that the Weibull stress scale parameter, σ_u, increases with temperature to reflect the increasing microscale toughness of ferritic steels caused by local events that include plastic shielding of microcracks, microcrack blunting, and microcrack arrest. The Weibull modulus, m, then characterizes the temperature invariant, random distribution of microcrack sizes in the material. Direct calibration of σ_u values at temperatures over the DBT region requires extensive sets of fracture toughness values. A more practical approach developed here utilizes the so-called Master Curve standardized in ASTM Test Method E1921-02 to provide the needed temperature vs. toughness dependence for a material using a minimum number of fracture tests conducted at one temperature. The calibration procedure then selects σ_u values that force the Weibull stress model to predict the Master Curve temperature dependence of K_Jc values for the material. At temperatures in mid-to-upper transition, the process becomes more complex as fracture test specimens undergo gradual constraint loss and the idealized conditions of high-constraint, small-scale yielding assumed in E1921-02 gradually degenerate. The paper develops the σ_u calibration process to incorporate these effects in addition to consideration of threshold toughness effects and the testing of fracture specimens with varying crack-front lengths. Initial illustrations of the calibration process for simpler conditions, i.e. 1T crack-front lengths, use the temperature dependent flow properties and a range of toughness levels for an A533B pressure vessel steel. Then using the extensive fracture toughness data sets for an A508 pressure vessel steel generated recently by Faleskog et al. [Engng. Fract. Mech., in press], the paper concludes with calibrations of both m and σ_u over the DBT region and assessments of the Master Curve calibration approach developed here.     

Move limits definition in structural optimization with sequential linear programming. Part II: Numerical examples

A variety of numerical methods have been proposed in literature in purpose to deal with the complexity and non-linearity of structural optimization problems. In practical design, sequential linear programming (SLP) is very popular because of its inherent simplicity and because linear solvers (e.g. Simplex) are easily available. However, SLP performance is sensitive to the definition of proper move limits for the design variables which task itself often involves considerable heuristics. This research presents a new SLP algorithm (LESLP) that implements an advanced technique for defining the move limits. The linearization error sequential linear programming (LESLP) algorithm is formulated so to overcome the traditional limitations of the SLP method. In a companion paper [Comput. Struct. 81 (2003) 197] the basics of the LESLP formulation along with a guide to programming are provided.The new algorithm is successfully tested in weight minimisation problems of truss structures with up to hundreds of design variables and thousands of constraints: sizing and configuration problems are considered. Optimization problems of non-truss structures are also presented. The numerical efficiency, advantages and drawbacks of LESLP are discussed and compared to those of other SLP algorithms recently published or implemented in commercial software packages.     

基于二维灰度图象的道路区域获取方法

针对二维灰度图象给出了一种获取道路区域的方法。首先基于边缘特性利用灰度直方图把初始生长点限定在道路区以内,然后确定区域生长阈值,以待生长点与已生长邻点的灰度接近性和已生成区域的均值接近性作为生长准则,形成准确的道路区域。