查询词典 algorithms

There has been considerable research in the field of stereo vision, and most stereo vision algorithms have been developed on the basis of regular lens camera model. However, such algorithms are unable to be used in the field where the clearance between the object and lens is minimal.

目前的双目或多目立体视觉方法基本上是在常规镜头摄象机成像模型的基础上发展起来的，但这些方法无法直接用于近距离大视场立体视觉感知的应用场合。

For regular topology surface, intermediate surface local texture mapping are used; for irregular topology surface, subdivision surface local texture mapping are used. So different topology surface using different texture mapping methods can solve the contradiction between texture effect and processing speed when they are dealt with by single method. The three judge conditions of texture distortion in the process of mapping onto the sphere surface are discussed and the exiting algorithms of mapping are analyzed, the new algorithms of mapping onto the sphere surface are presented. In order to enhance the realistic of graphics, a new function of wood texture is presented. OpenGL is also introduced in this papers and some realization examples of texture mapping based on OpenGL under the environment of VC++ and POV-RAY are presented in some chapters.

对于简单拓扑表面，利用改进的两步纹理映射，对于复杂拓扑表面利用新的面片分割算法，初步解决了颜色纹理中速度与走样之间的矛盾；通过对球面纹理时纹理不变形的三个判断条件的研究，分析了现在各种球面纹理映射算法，并在此基础上提出了一种基于等面积的半球面纹理映射算法，明显减少了球面纹理的纹理变形；针对现有的木纹纹理中，两种表示木纹的颜色差距表现出的颜色突变问题进行研究，提出对这两种颜色进行线性插值或二次插值，使生成的木纹纹理颜色更加连续，更加逼真；在VC++环境下，通过调用Open GL的图形库及在Pov-Ray软件环境下，实现了参数化纹理、球面纹理、凹凸纹理及木纹纹理等各种算法。

For these two algorithms, we present formal analysis of optimality, and experimental results show that the algorithms are efficient.

文中给出了两算法优化性能的分析，实验结果也显示这些算法的效果是明显而有效的。

Designing and implementing a prototype of optimizing compiler for URPR-1 which adopts the two-level software pipelining technique. Experiments show that URPR-1 and its compiler can obtain more than 5 times speedup for some typical signal and image processing algorithms and two orders-of-magnitude speedup for inner loops of these algorithms.

初步实验表明，URPR-1体系结构和它的优化编译器相互配合对信号和图象处理领域中的典型算法可以获得5倍以上的加速比，对这些算法的内层循环则可获得两个数量级的加速比。

Three global fraction orthogonal polynomials algorithms were reviewed in this paper,and their precision,efficiency and sensitivity to the testing noise were compared.Finally,based on the above comparison,the problems and future research direction of the algorithms were discussed.

本文系统的介绍了整体正交多项式识别算法发展过程中出现的三种方法，同时从识别精度、对噪声敏感程度和识别效率三个方面对方法进行详细的比较和评价，并在此基础上进一步指出了方法存在的问题和未来的研究方向。

In the parallel program, the in-core and out-of-core algorithms can be identified and chosen to use automatically. So the algorithms are united.(3) The re-orthonormalization technique is used to solve the loss of orthonormality of basis vector during the classical GMRES method solving large-scale three dimensional elasto-static BE problems.

3采用重正交技术解决了经典GMRES方法在求解大型三维弹性静力边界元问题过程中出现的基向量正交性丧失的问题；结合边界元基本解和系数矩阵的特点以及并行化的要求，研究了几种预条件技术；并以重正交技术和预条件技术实现了经典GMRES的实用化。

It is focused on the high efficiency clustering algorithms, outliner detecting algorithms, clustering result presentation methods, and so on. It is the basic work to define the similarity for high-dimensional data objects.

本文研究了高维聚类分析的关键技术，主要工作有：1、针对高维数据空间下聚类簇的分布特点，改进了一种基于子空间的映射聚类算法。

Theoretical and experimental results show that our algorithms outperform the existing algorithms in the STWin framework.

理论和实验结果表明ECSTA算法优于现有的STWin访框架下的算法。

This paper presents an evaluation method based on psychological scale. Rank-order method or method of paired comparison is adopted to sort the images retrieved by different algorithms, then an ordinal scale is obtained. With presupposition of normal distribution, an equal interval scale is also obtained though POZ conversion. Thus different algorithms can be compared on these scales.

本文提出基于心理量表的情感图像检索性能评价方法，采用等级排列法或对偶比较法对不同算法检索出的图像进行排序，建立顺序量表，并在样本满足正态分布假设的前提下，将顺序量表转化成等距量表，以比较检索算法的优劣。

The system consists of two modules: visual clustering analysis and visual data classifying, with respect to clustering algorithms and classifying algorithms, in which the visual technology of parallel coordinates is combined to obtain better visual data mining effects.

将聚类算法和分类算法与平行坐标的可视化技术结合起来，达到更好的可视化数据挖掘效果。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。