getpacket.wst2D(wavethresh)
getpacket.wst2D()所属R语言包：wavethresh
Get packet of coefficients from a two-dimensional non-decimated wavelet object (wst2D).
获取的系数的数据包从一个二维非抽取小波的对象（wst2D）。
译者：生物统计家园网 机器人LoveR
描述----------Description----------
This function extracts and returns a packet of coefficients from a two-dimensional non-decimated wavelet (wst2D) object.
此功能提取物并返回的系数的数据包从一个二维非抽取小波（wst2D）对象。
用法----------Usage----------
## S3 method for class 'wst2D':
getpacket(wst2D, level, index, type="S", Ccode=TRUE, ...)
参数----------Arguments----------
参数：wst2D
2D non-decimated wavelet object from which you wish to extract a packet from.
二维非抽取小波对象，该对象要提取数据包从。
参数：level
The resolution level of the coefficients that you wish to extract. Can range from 0 to nlevels(wpst)-1.
分辨率级别的系数要提取。范围可以从0到nlevels（WPST）-1。
参数：index
The index number within the resolution level of the packet of coefficients that you wish to extract. Index is a base-4 number which is r digits long. Each digit can be 0, 1, 2 or 3 corresponding to no shifts, horizontal shift, vertical shift or horizontal and vertical shifts. The number r indicates the depth of the resolution level from the data resolution i.e. where r = nlevels - level.
该指数在分辨率级别的数据包要提取的系数。指数是一个基4号为数字。每个数字可以是0，1，2或3，对应于没有移位，水平移位，垂直移位或水平和垂直偏移。数r表示从数据的分辨率，即r = nlevels - level的分辨率等级的深度。
Where there is a string of more than one digit the left most digits correspond to finest scale shift selection, the right most digits to the coarser scales (I think).
凡有多于一个数字最左边的数字对应于最优秀的大规模转移的选择，最右边的数字粗糙的尺度是一个字符串（我认为）。
参数：type
This is a one letter character string: one of "S", "H", "V" or "D" for the smooth coefficients, horizontal, vertical or diagonal detail.
这是的一个字母的字符串：1的“S”，“H”，“V”或“D”表示的平滑系数，水平，垂直或对角细节。
参数：Ccode
If T then fast C code is used to obtain the packet, otherwise slow SPlus code is used. Unless you have some special reason always use the C code (and leave the argument at its default).
如果T然后快速的C代码是用来获取数据包，否则慢的S-PLUS代码使用。除非你有一些特殊的原因，一直使用C代码（并留在其默认的参数）。
参数：...
any other arguments
任何其他参数
Details
详细信息----------Details----------
The wst2D function creates a wst2D class object. Starting with a smooth the operators H, G, GS and HS (where G, H are the usual Mallat operators and S is the shift-by-one operator) are operated first on the rows and then the columns: i.e. so each of the operators HH, HG, GH, GG, HSH, HSG, GSH, GSG HHS, GHS, HGS, GGS HSHS, HSGS, GSHS and GSGS are applied. Then the same collection of operators is applied to all the derived smooths, i.e. HH, HSH, HHS and HSHS.
wst2D函数创建一个wst2D类对象。开始的具有光滑的运营商H，G，GS和HS（式中，G，H是通常的Mallat运营商和S是变速通过一个操作员）操作第一的行和列上：即，使各应用运营商HH，HG，GH，GG，香港上海大酒店，HSG），谷胱甘肽过氧化物酶（GSH，GSG HHS，GHS，HGS，GGS HSHS，HSGS，GSHS和GSGS。然后，运营商在同一集合的应用到所有派生平滑，HH，香港上海大酒店，HHS和HSHS。
So the next level is obtained from the previous level with basically HH, HG, GH and GG but with extra shifts in the horizontal, vertical and horizontal and vertical directions. The index provides a way to enumerate the paths through this tree where each smooth has 4 children and indexed by a number between 0 and 3.
因此，下一级得到从以前的水平基本上是HH，HG，GH和GG，但额外的在水平，垂直和水平及垂直方向上的变化。该指数提供了一种方法来枚举的路径，通过这棵树，每光滑有4个孩子和索引之间的一个数字0和3。
Each of the 4 children has 4 components: a smooth, horizontal, vertical and diagonal detail, much in the same way as for the Mallat 2D wavelet transform implemented in the WaveThresh function imwd.
4的儿童中的每一个都具有4部分组成：一个平滑，水平，垂直和对角线细节，许多相同的方式的Mallat二维小波变换中实现的WaveThresh功能imwd。
值----------Value----------
A matrix containing the packet of the 2D non-decimated wavelet coefficients that you require.
一个矩阵包含的二维非抽取小波系数的，你需要的数据包。
RELEASE----------RELEASE----------
Version 3.9 Copyright Guy Nason 1998
版本3.9版权所有1998年盖利晨
（作者）----------Author(s)----------
G P Nason
参见----------See Also----------
putpacket.wst2D, wst2D, wst2D.object.
putpacket.wst2D，wst2D，wst2D.object。
实例----------Examples----------
#[]
# Create a random image. [创建一个随机的图像。]
#[]
myrand <- matrix(rnorm(16), nrow=4, ncol=4)
#myrand[myrand]
# [,1] [,2] [,3] [,4] [[1] [2] [3] [4]]
#[1,] 0.01692807 0.1400891 -0.38225727 0.3372708[[1，] 0.01692807 0.1400891 -0.38225727 0.3372708]
#[2,] -0.79799841 -0.3306080 1.59789958 -1.0606204[[2] -0.79799841 -0.3306080 1.59789958 -1.0606204]
#[3,] 0.29151629 -0.2028172 -0.02346776 0.5833292[[3] 0.29151629 -0.2028172 -0.02346776 0.5833292]
#[4,] -2.21505532 -0.3591296 -0.39354119 0.6147043[[4] -2.21505532 -0.3591296 -0.39354119 0.6147043]
#[]
# Do the 2D non-decimated wavelet transform[做二维非抽取小波变换]
#[]
myrwst2D <- wst2D(myrand)
#[]
# Let's access the finest scale detail, not shifted in the vertical[让我们进入规模最优秀的细节，而不是在垂直方向上移动]
# direction.[的方向发展。]
#[]
getpacket(myrwst2D, nlevels(myrwst2D)-1, index=0, type="V")
# [,1] [,2] [[1] [2]]
#[1,] -0.1626819 -1.3244064[[1，] -0.1626819 -1.3244064]
#[]
# Compare this to the ordinary 2D DWT for the vertical detail at this[与此相比，普通的2D DWT的垂直细节]
# resolution level[分辨率级别]
imwd(myrand)[[lt.to.name( 1, "DC")]]
#[1] -0.1626819 -1.3244064 1.4113247 -0.7383336[[1] -0.1626819 -1.3244064 1.4113247 -0.7383336]
#[]
# The same numbers but they're not in matrix format because[同样的数字，但他们不以矩阵格式，因为]
# imwd returns vectors not matrices.[imwd回报向量矩阵。]
#[]
# Now back to the wst2D object. Let's[现在回到的wst2D对象。让我们]
# extract vertical detail again at level 1 but this time the horizontally[提取垂直细节，再次在第1级，但这段时间的水平]
# shifted data.[转移数据。]
#[]
getpacket(myrwst2D, level=1, index=1, type="V")
# [,1] [,2] [[1] [2]]
#[1,] -0.5984427 0.2599445[[1，] -0.5984427 0.2599445]
#[2,] -0.6502002 1.8027955[[2，] -0.6502002 1.8027955]
#[]
# So, yes, different data. Now how about at a deeper resolution level.[所以，是的，不同的数据。现在怎么样在更深的分辨率水平。]
# Lets have a horizontal shift, as before, for the level 1 but follow it[让有一个水平移位和以前一样，为在第1级，但跟随它]
# with a diagonal shift and this time extract the smooth component:[一个对角线的转变，这一次提取的平滑分量：]
#[]
getpacket(myrwst2D, level=0, index=13, type="S")
# [,1] [[1]]
#[1,] -0.5459394[[1，] -0.5459394]
#[]
# Of course, only one number because this is at level 0[当然，只有一个号码，因为这是在级别0]
转载请注明:出自 生物统计家园网(http://www.biostatistic.net)。
注：
注1：为了方便大家学习，本文档为生物统计家园网机器人LoveR翻译而成，仅供个人R语言学习参考使用，生物统计家园保留版权。
注2：由于是机器人自动翻译，难免有不准确之处，使用时仔细对照中、英文内容进行反复理解，可以帮助R语言的学习。
注3：如遇到不准确之处，请在本贴的后面进行回帖，我们会逐渐进行修订。
|
发表于 2012-10-1 17:20:53
