三维服装参数化设计技术

通过拓展传统参数化,在三维服装设计中引入曲线曲面作为参数化的基本元素,提出包括模型构建、模型驱动与模型联动的三维服装参数化设计技术框架及其实现.利用特征信息分层次构建服装模型,得到具有非精确表达的服装几何模型;利用混合维度输入信息,对服装多层次几何模型进行多因素驱动;基于模型层次关联信息,对服装模型进行多层次联动和混合维度联动,并能得到系列化产品.相应的实例表明,文中方法能有效地实现人体、衣片和裁片的参数化设计,从而为三维服装大规模定制提供设计方法与技术支持.     

产品广义参数化设计的研究

分析了参数化设计的研究现状,提出了面向设计周期的产品广义参数化设计的概念,建立了从产品概念设计到详细设计的广义参数化信息模型,定义不同层次的参数映射和约束表达方法,通过主参数来封装产品几何信息和非几何信息,从而最终实现产品级、部件级与零件的参数化设计。     

CAD开发中图形参数化的实现途径探

CAD是一种具有强大功能的计算机辅助设计技术的统称,其技术的核心就是利用计算机的硬件功能对产品进行高速度的设计,在设计的过程中为了提高设计效率,技术人员引入了参数化的模式,即在产品形状相对固定的情况下,特定参数的变化来完成某些设计内容,简单的看就是改变参数而产品的几何尺寸就会随之改变,而达到设计新产品的目的。参数化的实现途径主要是利用数据库和相应的模型建立来达成,即通过数据库和CAD之间的连接来构建参数化的途径。     

基于ACIS/HOOPS的特征造型系统设计与实现

为了探索CAD/CAM领域中的特征造型关键技术,研发一个特征造型系统。该系统分别以ACIS和HOOPS为几何造型引擎和渲染引擎,能够实现常用特征的创建、特征历史的管理、基本约束的管理和特征模型与几何模型的存取等功能,并且支持一些自由曲面特征造型以及参数化设计。为了维护用户的隐含设计意图,系统采用了一种新的拓扑元素命名与辨识机制。最后,给出了使用该系统进行特征建模、自由曲面特征参数化设计以及设计意图有效维护等实例,验证了本系统的可行性和有效性。     

改进的灰色模型与ARMA模型的股指预测

当前基于灰色GM（1,1）模型和ARMA模型的组合模型GM-ARMA模型存在着2点不足：一是由于GM（1,1）模型不是最优的,导致了GM-ARMA模型也不是最优的;二是GM-ARMA模型并没有恰当地结合2个子模型,这也导致了GM-ARMA模型不是最优的.为此,首先引入数据维度参数和白化背景值的系数2个参数来改进GM（1,1）模型,然后同时优化ARMA模型中的P、Q2个参数来改进GM-ARMA模型,称新的模型为RevisedGM-ARMA（RGM-ARMA）模型.实例证明RGM-ARMA的误差小于ARIMA和GM-ARMA模型,并且为组合模型的建立提供了新的思路.     

带形状参数的指数均匀B样条模型

本文给出了k(k≥2)阶带形状参数指数多项式的均匀B样条模型.该类模型具有很多与B样条模型相同的性质,并且具有一个可调节的形状参数.由该模型构造的曲线,通过改变形状参数的取值,可以调整曲线接近其控制多边形的程度.该模型可以应用于CAD/CAM领域,作为几何造型一种新的有效模型.     

带多形状参数的三角多项式均匀 B样条曲线曲面

由于在进行几何外形设计时对曲线曲面的局部调控能力要求越来越高,为了给设计者们提供更丰富的方法,利用分段积分的思想构造了一类带多个形状参数的三角多项式均匀B样条曲线曲面,并讨论了这类曲线曲面所具有的重要性质.通过改变形状参数的取值来整体或局部调控曲线曲面形状,随着曲线阶数的升高扩展形状参数的取值范围;通过公式推导给出了曲...     

基于切向约束构造复合二次B样条插值曲线

该文提出一种构造二次B样条插值曲线的新方法,包括新的参数化方法和新的插值方法.新参数化方法中,相邻曲线段的连接处与插值点相一致,以插值点的切向作为约束,利用二次B样条曲线本身的几何性质进行参数化,使曲线在每个插值点上都满足指定的切向,可以直观地控制插值曲线的形状以达到预期效果,参数化方法稳定,不必解方程组.在新参数化方法的基础上进一步提出了分段构造的思想,将形状不好的段分成多段构造,除插值点的切向外还留有其他的自由度进一步直观调控曲线的形状,使得二次B样条插值曲线的形状更自然.新方法对于数据点的改变具有良好的局部性.实例表明该方法是有效的.     

一种新的活动轮廓模型--S-L模型

活动轮廓模型用于图像分割一般分为两种基于参数的模型和基于几何特征的模型.Snake模型可以快速地分割目标,但不能处理拓扑结构复杂的情况且对初值位置过于敏感.水平集模型具有拓扑可变性,但其时间效率较低,在分析这两种模型优缺点的基础上,提出了一种新的活动轮廓模型,该模型兼具有上述两种方法的优点快速性、拓扑可变性.在模型中用Snake模型的能量方程控制曲线的演化并提出一种基于水平集思想的符号表法来改变演化过程中曲线的拓扑结构.为了降低噪音的影响,用区域信息构造新的外力,在外力的作用下可以使初始曲线有更大的选择空间.对左心室MR图像的分割实验结果表明,该模型得到的分割结果与Level Set模型相似,但所用时间远比Level Set模型少.     

几何偏微分方程和离散曲面设计

使用若干个几何本质的曲率驱动的偏微分方程来构造符合指定C0或C1边界条件的三边曲面片和四边曲面片,这些方程的数值解由所涉及的微分几何算子的离散化来得到,微分几何算子的离散化则源于参数逼近.所构造的曲面片满足某些特定的几何偏微分方程,故具有理想的形状,将这些曲面片组装起来便构造出复杂的几何模型.通过反复的子分和演化,得到几何模型的多尺度表示.     

Numeric and curve parameters for freeform surface feature models

The success of feature modelling systems depends to a large extent on offering user-friendly ways for specifying and varying feature models. In this paper, several new facilities are introduced for this.Numeric and curve parameters are presented for specification and variation of freeform surface feature models. Several properties of the parameters can be defined, such as the direction of evaluation for a curve parameter. In addition to parameters present in features, other parameters can be added to a feature model, including intersection curves between features.To provide these facilities, a new three-level structure is introduced, which includes a partially evaluated model that can capture new generic model entities, including feature intersection curves. The work has been implemented in a prototype surface feature modelling system with functionality for feature class definition, feature model creation, and advanced feature model variation through the new types of parameters.     

基于视频的车辆几何参数提取与车型分类

针对车辆自动分类需求,通过引入地理信息系统空间形态的描述方法,给出车辆侧面图像的几何信息描述参数和计算方法,设计用于车辆分类的几何特征参数及其组合参数,基于侧视轮廓的几何特征,提出车辆轮轴参数的自动提取方法,设计一个基于几何形态参数的车型自动分类器。实验证明该方案的有效性。     

Dynamic interfacing of applications to geometric modeling services via modeler neutral protocol

An architecture was developed for standardizing communications between geometric modeling core systems and applications that require geometric modeling services, such as feature modelers, feature recognition and process planning systems. Since geometric kernels differ in the functionality and applications vary in the level of geometric services required, a multi-layered communication architecture was developed and implemented. The methodology is analogous to the X-Window standard for graphics display, albeit the domain is different. At the lowest level is a library of classes, named Geo-modeler, whose methods offer low level solid modeling services. The member methods of these classes are translated into geometric modeler specific commands. If and when STEP extends to this domain, these specific calls could be replaced by a standard dynamic protocol (such as standard SDAI calls) and the translation done by the vendor within his geometry kernel. On top of Geo-modeler there is another layer, called Geo-widgets, which is written entirely using Geo-modeler functions. At the highest level the Geo-tools are functions used commonly by generic applications. Feature applications can choose to use the library at any level. The intermediate layers (geo-tools, Geowidgets) make all geometry kernels appear to have the same functionality to applications, thus creating a ‘plug compatible’ environment between applications and the geometric modeling kernel. The architecture has been tested with ACIS^® and Parasolid^® for a variety of design and manufacturing applications.     

High-level feature recognition using feature relationship graphs

In this paper, we present a method to extract user-specific features from common features. This contrasts with other approaches which work directly off B-rep geometric models. Here, user-specific features are called high-level features which are a set of common features combined in a user-specific manner. A feature relationship graph is used to organize common features in a part and to define high-level feature patterns. The research presented in this paper focuses mainly on feature relationship graph construction and high-level feature recognition using subgraph isomorphic techniques.     

Enriched Latent Dirichlet Allocation for Sentiment Analysis

One of the main benefits of unsupervised learning is that there is no need for labelled data. As a method of this category, latent Dirichlet allocation (LDA) estimates the semantic relations between the words of the text effectively and can play an important role in solving various issues, including emotional analysis in combination with other parameters. In this study, three novel topic models called date sentiment LDA (DSLDA), author–date sentiment LDA (ADSLDA), and pack–author–date sentiment LDA (PADSLDA) are proposed. The proposed models extend LDA through some extra parameters such as date, author, helpfulness, sentiment, and subtopic. The proposed models use helpfulness in the Gibbs sampling algorithm. Helpfulness is a part of readers who found the review helpful. The proposed models divide the words into two categories: the words more affected by the distribution of subtopic and the words more affected by the main topic. In this study, a new concept called pack is introduced, and a new model called PADSLDA is proposed for sentiment analysis at pack level. The proposed models outperformed the baseline models because according to evaluations results, the extra parameters can appropriately affect the generating process of words in a review. Sentiment analysis at the document level, perplexity, and topic coherence are the main parameters used in the evaluations.     

一种遥感图象的混合多尺度Hurst参数分类模型

针对遥感图象分类较困难的问题,提出了一个遥感图象的分类模型－扩展的自相似模型（ESS）,该模型是一种广义的分形布朗模型（fBm),它的多尺度Hurst参数与粗糙度之间的是对应的,同时不必像分形维数那样要求粗工的尺度不变性,因而比fBm更接近于实际情况,另外,由于它的参数可以作为很好的分类特征,而且特征给数低,计算快,其方向性Hurst参数还描述了纹理在4个方向上的粗糙度,因此可将它们与灰度的均值和标准差一起作为一组特征,来构造一个混合多尺度Hurst参数分类模型,将其用于卫星遥感图象分类,获得了较高的分类正确率。     

基于边缘几何特征的图像精确匹配方法

提出一组快速高精度计算切线斜率的五点公式，用以估计图像边缘曲线的角度特征，并利用角度直方图估计图像几何变换的旋转参数，实现具有大旋转差异图像间的粗匹配．在进行角度补偿后，利用灰度互相关判据搜索匹配点对，计算出几何变换参数，实现较高精度的旋转和平移校正，最后用松弛迭代法完成图像的精确匹配．与基于小波方向角特征的匹配方法相比，文中方法利用图像中主要的边缘信息实施匹配，具有较好的鲁捧性，可成功实现对各类具有较大相关程度图像间的精确匹配，对图形匹配也具有重要意义．     

Geometric invariant features in the Radon transform domain for near-duplicate image detection

Radon transform has been widely used in content-based image representation due to its excellent geometric properties. In this paper, we propose a family of geometric invariant features based on Radon transform for near-duplicate image detection. According to the theoretical analysis between geometric operations (translation, scaling, and rotation) and Radon transform, we present a geometric invariant feature model. Based on the feature model, we developed four kinds of geometric invariant features. In addition, a uniform sampling technique is introduced to combine different features. The comprehensive performance of the combined feature is better than that of a single one. Extensive experiments show that the proposed features are robust, not only to rotation and scaling, but also to other operations, such as compression, noise contamination, blurring, illumination modification, cropping, etc., and achieve strong competitive performance compared with the state-of-the-art image features.     

Performance modeling of a two-echelon supply chain under different levels of upstream inventory information sharing

The advancement in information technology has facilitated the sharing of information in supply chain networks (SCNs), resulting in effective management of inventory and storage capacity. In this paper, our focus is on upstream inventory information sharing. Existing analytical performance evaluation models of SCNs are not capable of assessing the impact of inventory information sharing. To address this need, we develop performance evaluation models of SCNs that explicitly consider production capacity, inventory related decisions, variability, transit delays and inventory information sharing in a unified manner. We employ a two-echelon SCN configuration with two retail stores and two production facilities as a test bed. The retail stores have inventory information from the production facilities. We model three levels of inventory information sharing in our study; the information shared ranges from the stock-out information at the lowest level to inventory and backorder level information at the highest level. We develop analytical models first for Poisson arrivals and exponential processing times under all levels of inventory information sharing. We extend these models to general inter-arrival and processing time distributions and subsequently include transit delays between the production facilities and the retail stores. We demonstrate the performance prediction capability of the analytical models developed via extensive numerical experimentation.     

新颖的判别性特征选择方法

作为数据预处理的一种常用的手段,特征选择不仅能够提高分类器的分类性能,而且能增加对分类结果的解释性。针对基于稀疏学习的特征选择方法有时会忽略一些有用的判别信息而影响分类性能的问题,提出了一种新的判别性特征选择方法——D-LASSO,用于选择出更具有判别力的特征。首先D-LASSO模型包含一个L₁-范式正则化项,用于产生一个稀疏解;其次,为了诱导出更具有判别力的特征,模型中增加了一个新的判别性正则化项,用于保留同类样本以及不同类样本之间几何分布信息,用于诱导出更具有判别力的特征。在一系列Benchmark数据集上的实验结果表明,与已有方法相比较,D-LASSO不仅能进一步提高分类器的分类精度,而且对参数也较为鲁棒。