2) weight matrix
权值矩阵
1.
Algorithm weight matrixes accurately express the complex non-linear relationships of lattice parameters of the two phases with temperatures and phase compositions are obtained based on the back-propagation neural network modeling on the practical values of temperature, phase composition and lattice parameter in various superalloys.
基于对多种镍基高温合金在不同温度下两相成分及其点阵常数数据归一化处理的反向传播神经网络模拟,得到了可以比较准确表达两相点阵常数与相成分、温度间复杂非线性关系的权值矩阵,由此可对镍基合金在不同温度下γ’和γ相的点阵常数进行测算,并获得与X射线衍射法测定数据甚为接近的点阵常数数据,证实了该方法的可行性及准确性。
2.
The module has functions of creating weight matrix,global spatial autocorrelation analysis and local spatial autocorrelation analysis,and can be plugged into ArcGIS applications.
模块包括空间权值矩阵建立、全局空间自相关分析、局部空间自相关分析三方面的功能,并可以嵌入到A rcG IS系统中。
3) matrix of initial value
初值矩阵
1.
Through calculating matrix of initial values, based on original values and state equations of the circuit, and solving the state equations, complete responses (analytic solutions) of the circuit are given.
提出了一种计算线性电路状态方程解析解的新方法 :以电路的原始值和状态方程为基础计算出电路的初值矩阵并求解电路的状态方程 ,从而获得该电路的完全响应 (解析解
4) dual-value matrix
二值矩阵
1.
In the process of its realization, it is necessary to transform the colored graphics on CRT into dual-value matrix, and then to amplify it, furthermore, the amplified lattice of graphics can be shaped by using an appropriate shaping algorism, meanwhile, in order to shorten the matrix operation time and to reduce the internal memory, the shaped graphics should be cut apart.
实现过程中需要将屏幕彩色图形转换为二值矩阵,然后进行相应的放大处理,并可以使用适当的整形算法对放大后的图形点阵进行整形;同时,为了缩短矩阵运算时间,降低内存,要对整屏图形进行适当的分割分别处理。
5) difference matrix
差值矩阵
1.
Fractal image encoding based on distance between difference matrix of blocks;
基于域块差值矩阵距离的分形图像编码
2.
Reversible data hiding algorithm based on difference matrix of the original image is proposed in this paper, the property of central distributing of the elements in the difference matrix is used to generate the embedding space, the secret information is embedded through modifying the elements of the difference matrix, and this can be mapped to the modification of the pixels of the carrier.
本文提出一种基于宿主图像差值矩阵的可逆信息隐藏算法,利用差值矩阵中元素分布集中的特性,生成秘密信息的嵌入空间,完成图像预处理,再通过修改差值矩阵元进而映射修改宿主图像的像素,完成秘密信息的嵌入,在接收端可以提取秘密信息,同时能够完全恢复宿主图像。
3.
It uses the property that the difference between adjacent pixels is always small and utilizes the difference matrix to do the original image preprocessing,amends and maps the margin matrix to the original image to realize information hiding.
该算法利用图像相邻像素之间差异较小的特性,通过差值矩阵对原始图像进行预处理,修改差值矩阵并映射到宿主图像中实现信息隐藏。
6) threshold matrix
阈值矩阵
1.
In consideration without increase of computation complexity, we propose an algorithm for gray image segmentation by creating a threshold matrix obtained by global threshold and dynamic threshold that are the FCM clus.
模糊C均值算法用于灰度图像分割是一种非监督模糊聚类后再标定的过程,该文在不明显增加运算量的前提下,利用模糊C均值自动聚类的功能分别得到全局阈值和动态阈值,完成对阈值矩阵的构造和图像的分割。
补充资料:单值矩阵
单值矩阵
monodromy matrix
单值鹭t黑嘿暮黔黑瀑巍常薇分方程又”A(t)x,t6R,x‘R”在零点处正规化的基本矩阵(和」、da此幻ta]袱亩认)X(亡)当:二。时的值;其中A(0是。周期矩阵,它在R的每一个紧区间上是可和的.刃一B kn“””~柳
说明:补充资料仅用于学习参考,请勿用于其它任何用途。
参考词条