面向对象可重用部件库的管理技术

本文讨论了面向对象可重用部件库管理中的几个技术问题，重点阐述了部件的浏览、检索和匹配技术。     

基于ICA/SVM的混合学习方法及在人脸检测中的应用

从基于样本学习的方法框架出发,该文提出了一种基于独立元分析和支持向量机的混合的学习方案,并用于人脸检测中。该方法通过独立元分析方法进行特征提取,然后采用SVM进行分类。该文做了大量的实验,以确定如独立元个数等参数问题对该分类器的影响,并与单独的支持矢量机方法、其它的人脸检测方法进行了比较分析。实验结果表明,该文的方法具有较好的检测效果,是一种很有效的方法。     

基于进程组构件模型的软件体系结构设计方法

软件工程的有效性在于进行体系化或系统化的设计，软件体系结构的好坏对系统的成败起着举足轻重的作用。本文针对并发系统，结合面向对象、构件以及软件复用的思想，提出了一种基于进程组构件模型的软件体系结构设计方法。最后介绍了采用该方法设计的iCALL系统。     

Quasilinearized Regression

Estimation in nonlinear regression usually requires a considerable amount of computational work. If it is possible to linearize the original formulas, the linear regression is sometimes applied to transformed linear relations. The method of quasilinearized regression (first order approximation of nonlinear regression), comparable to linear regression, with respect to computational effort, is described in this paper. This method gives substantially better results than linear regression of linearized formula.     

Computing Maximum Likelihood Estimates for the Mixed A.O.V. Model Using the W Transformation

The W transformation, a matrix transformation, is developed and applied for the mixed analysis of variance model to compute maximum likelihood estimates of the variance components and fixed parameters. This transformation not only eliminates the need for the explicit computation of the n × n inverse matrix H^?l but permits handling the iterative calculations such that they do not depend upon n (the number of observations) in any way. Although not wedded to a particular numerical method, the W transformation is implemented in conjunction with a modified Newton-Raphson method in which variance components are restricted to being non-negative.     

A Method of Estimating Variance Components in Unbalanced Designs

Another method of estimating variance components in the unbalanced case is presented for a general model with unbalanced population structure and unequal variances over the population. The method utilizes unweighted means and can be shown to give unbiased estimates when the numbers of observations are sampled and certain to be > 1. When the design is balanced the estimators coincide with the usual anova-estimators. The construction of estimators is exemplified for four situations with factors crossed and nested in different ways.     

Graduated driver licensing program component calibrations and their association with fatal crash involvement

Graduated driver licensing (GDL) programs are specialized licensing systems for beginner drivers adopted in all U.S. states, Australia, Canada, and New Zealand. GDL programs reduce novice drivers’ exposure to high-risk driving situations while they gain driving experience. Several studies document the success of GDL programs overall in reducing young teen crash rates. However, little is known about which specific components of these programs (e.g., nighttime driving restrictions) and which calibrations of these components (e.g., 10 PM, 11 PM, 12 AM, or 1 AM), are associated with the largest crash reductions. The goal of this study was to identify the GDL component calibrations associated with the largest reductions in fatal crash involvements for 16–17-year-olds. Driver fatal crash involvements for all U.S. states from 1986 to 2007 were analyzed using Poisson regression models to estimate the association of various GDL component calibrations with 16- and 17-year-old driver fatal crash incidence, after adjusting for potential confounders. We found clear evidence that (a) a minimum learner permit holding period of 9–12 months and (b) a passenger restriction allowing only one teen passenger for 6 months or longer are the calibrations for the learner permit holding period and passenger restriction components associated with the largest reductions in 16–17-year-old driver fatal crash involvements. Additionally, the data suggest that (a) disallowing learner driving until age 16, (b) disallowing intermediate licensure until age 16½ to 17, and (c) a nighttime driving restriction starting at 10 PM or earlier are the calibrations for these components associated with the largest reductions in 16-year-old driver fatal crashes. There was no clear evidence to support particular calibrations for supervised driving hours or unrestricted license ages.     

Electronic nose and GC–MS analysis of volatile compounds in Tuber magnatum Pico: Evaluation of different storage conditions

The characteristic aromatic composition of white truffles (Tuber magnatum Pico) determines its culinary and commercial value. However modifications of truffle organoleptic proprieties occur during preservation. A study of headspace of white truffles by using Electronic nose (E-nose), gas chromatography–mass spectrometry (GC–MS) and sensory analyses was performed. Truffles were stored at different conditions for 7 days: +4 and +8 °C wrapped in blotting paper or covered by rice or none of the above. Headspace E-nose measurements and sensory analyses were performed each day. Statistical multivariate analysis of the data showed the capability of E-nose to predict sensorial analysis scores and to monitor aroma profile changes during storage. Truffle’s volatile molecules were also extracted by headspace solid phase microextraction technique and separated and identified by GC–MS. Partial Components Analysis of data was performed. E-nose and GC–MS results were in agreement and showed that truffle storage in paper at +8 °C seemed to be the best storage condition.     

A model of domain-polymorph component for heterogeneous system design

Heterogeneous systems mix different technical domains such as signal processing, analog and digital electronics, software, telecommunication protocols, etc. Heterogeneous systems are composed of subsystems that are designed using different models of computation (MoC). These MoCs are the laws that govern the interactions of the components of a subsystem. The design of heterogeneous systems includes the design of each part of the system according to its specific MoC, and the connection of the parts in order to build the model representing the system. Indeed, this model allows the MoCs that govern different parts of system to coexist and interact.To be able to use a component which is specified according to a given MoC, under other, different MoCs, we can use either a hierarchical or a non-hierarchical approach, or we can build domain-specific components (DSC). However, these solutions present several disadvantages. This paper presents a new model of component, called domain-polymorph component (DPC). Such a component is atomic and is able to execute its core behavior, specified under a given MoC, under different host MoCs. This approach is not a competitor to the approaches above but is complementary.     

H型钢梁多机器人焊接工作站设计与仿真

针对H型钢梁人工焊接效率低、成本高、劳动强度大且对其自动化生产的研究较少的情况,通过分析其焊接过程,设计一种由两台焊接机器人和一台搬运机器人组成的自动化焊接工作站。利用SolidWorks设计了扩大机器人工作范围的第七轴和保证机器人运行轨迹与焊缝不发生偏离的焊缝自动跟踪系统,在数字化仿真软件Visual Components中搭建机器人焊接工作站布局,采用遗传算法规划最优焊接路径,利用仿真软件完成了编程与仿真。仿真结果表明所建立工作站布局的正确性。