ANUSPLIN3.1插值软件_anusplin插值资源-CSDN文库

共22个文件

exe：14个

doc：5个

eer：1个

5星 · 超过95%的资源需积分: 49 60 浏览量 2011-06-01 11:32:50 上传评论 8 收藏 1.54MB RAR 举报

ANUSPLIN是一款强大的地理信息系统（GIS）中的插值软件，专用于空间数据的光滑插值和预测。这款工具在3.1版本中提供了一系列先进的统计方法和技术，以处理复杂的空间数据模式，使得用户能够更准确地理解地理分布特征。 1. **插值方法**： ANUSPLIN的核心功能是基于自然基函数的光滑插值，这种插值方法通过构建高度灵活的数学模型来逼近原始数据点，同时保持插值结果的连续性和光滑性。这种方法在处理地形、气候、地质等空间数据时尤其有效。 2. **自适应分段多项式**： ANUSPLIN使用的是自适应分段多项式，这意味着它可以根据数据的局部特征调整多项式的阶数，以适应不同区域的数据复杂性。这有助于避免过拟合或欠拟合问题，确保模型对数据变化的敏感度适中。 3. **空间预测**：除了插值，ANUSPLIN还能进行空间预测。通过对已有数据点的插值分析，可以预测在未采样点上的变量值，这对于环境科学、地球科学等领域进行空间扩展和预测分析极其有用。 4. **数据处理能力**： ANUSPLIN可以处理大量的观测数据，这对于大规模的空间数据分析而言至关重要。它可以有效地处理数以千计甚至更多的数据点，生成高分辨率的格网数据。 5. **精度与效率**： ANUSPLIN算法设计兼顾了精度和计算效率。虽然其插值模型复杂，但计算速度相对快速，适合在资源有限的环境下进行大规模数据处理。 6. **GIS集成**：作为一款GIS插件，ANUSPLIN可以无缝集成到各种地理信息系统软件中，如ArcGIS、QGIS等，方便用户在熟悉的环境中进行数据处理和分析。 7. **应用领域**： ANUSPLIN广泛应用于环境科学、气象学、水文学、地质学等多个领域。例如，它可以用来分析气候模型的输出，进行地形表面重建，或者预测地下水位的变化等。 8. **版本更新**： ANUSPLIN 3.1版可能包含了对前一版本的优化和新功能的添加，例如改进的算法性能、增强的用户界面或支持更多数据格式。不过，具体更新内容需要参照软件的官方文档或发布说明。 9. **用户支持**：使用ANUSPLIN时，用户通常可以期待详细的文档、教程和用户社区的支持，帮助解决在安装、操作和分析过程中遇到的问题。 10. **注意事项**：在使用ANUSPLIN时，用户需要注意正确处理数据的预处理，如数据清洗、坐标转换和缺失值处理。此外，选择合适的插值参数对结果的准确性有很大影响，需要根据实际数据特点进行调整。 ANUSPLIN 3.1版是地理空间数据处理的一个强大工具，通过其高效的插值和预测功能，为科研和实践提供了有力的支持。

资源推荐

资源详情

资源评论

收起资源包目录

ANUSPLIN.rar （22个子文件）

ANUSPLIN

AVGCVB.EXE 183KB

ERRGRD.EXE 341KB

LAPPNT.EXE 308KB

ERRPNT.EXE 312KB

DOC

SPLINAB.DOC 22KB

LAPPNT.DOC 3KB

ANUSPLIN.DOC 4KB

LAPGRD.DOC 7KB

SPLNAABB.DOC 6KB

README 188B

ADDNOT.EXE 78KB

SPLINB.EXE 365KB

DELNOT.EXE 73KB

LF90.EER 40KB

README.DOS 2KB

SPLINBB.EXE 365KB

SELNOT.EXE 305KB

AVGCVA.EXE 222KB

SPLINA.EXE 366KB

LAPGRD.EXE 338KB

SELNOTT.EXE 305KB

SPLINAA.EXE 367KB

Documentation for SPLINA and SPLINB VERSION 3.1

-----------------------------------------------

SPLINA is a FORTRAN program which fits a partial thin plate smoothing

spline to multi-variate noisy data.

SPLINB is an approximate version of SPLINA designed for larger data

sets. It fits a partial thin plate smoothing spline, with user defined

knots, to multi-variate noisy data.

Author: M.F.Hutchinson, CRES, Australian National University,

Canberra ACT 0200, Australia

Last revision: 11 November 1996.

Frequent users may wish to set an alias to simplify program access.

User directives are read from the standard input unit. Users are

advised to use a command file for these so that the output can be saved.

This output includes essential statistical information on the fitted

surfaces, including residuals and standard error estimates.

The command file should only contain the program directives. The program

is then executed by typing, for example,

splina < cmdfile > cmdfile.lis

The original thin plate (formerly Laplacian) surface fitting technique

was described by Wahba (1979), with modifications for larger data sets

due to Bates and Wahba (1982), Elden (1984) and Hutchinson (1984). The

extension to partial splines is based on Bates et al. (1987). This

allows the incorporation of parametric linear sub-models (or covariates)

into the fitted spline, in addition to the usual independent spline

variables. They are a robust way of allowing for such dependencies

provided a parametric form for this dependence can be determined. In

the limiting case of no independent spline variables (not currently

permitted), the procedure would become a simple multivariate linear

regression procedure. A comprehensive introduction to the technique of

thin plate smoothing splines, with various extensions, is given in Wahba

(1990). A brief overview of the basic theory and applications is given

in Hutchinson (1991a). More comprehensive discussion of the algorithms

and the associated statistical analyses are given in Hutchinson and

Gessler (1994) and Hutchinson (1993,1995).

Program Input

-------------

This includes:- the number of independent spline variables, the number

of independent covariates, the lower and upper limits for each

independent variable, the order of derivative to be minimized, the

number of surfaces, the data error standard deviation estimate for each

surface and input and output file specifications, including the knot

index file name in the case of SPLINB. Data points at positions which

lie outside the user supplied independent variable limits are rejected.

This may be useful as a way of fitting a surface to a subset of the data

without having to create a separate data file. It also gives a simple

check on the specified data format and the order of the independent

variables in the data file. An error in these specifications would be

indicated if fewer than the expected number data points are selected.

SPLINA may currently fit up to 12 surfaces to data at up to several

hundred sites. SPLINB, the approximate version of the thin plate smoothing

procedure, may fit up to 12 surfaces to data at up to a few thousand sites

using up to several hundred knots.

For SPLINA, the main storage requirements of the program are proportional

to the square of the number of data points and the processing time is

proportional to cube of the number of data points. For SPLINB, storage

requirements are proportional to the square of the number of knots and

processing time is approximately proportional to the cube of the number of

knots, since a singular value decomposition of a matrix of this order is

required. The required linear algebra routines are contained within the

double precision LINPACK library (Dongarra et al., 1979). The required

LINPACK routines can be supplied with ANUSPLIN.

SPLINA does not permit coincident data points. SPLINB does permit

coincident data points. Coincident knots are not permitted. Simplified

procedures for processing coincident data points are being developed.

Program Output

--------------

Summary statistics and a list of the 100 largest residuals, ranked in

descending order, are always written to the standard output unit.

Depending on values of the user directives, a list of the data and

fitted values with estimated standard errors may be written to output

and files containing the coefficients of the fitted surfaces and the

parameters used to calculate the optimum smoothing parameter may be

written. The surface coefficients are used to calculate values of the

fitted surface by other programs in ANUSPLIN, including LAPGRD and

LAPPNT. The ranked list of residuals is particularly useful in

detecting miscellaneous data errors. The list of data and fitted values

is also useful for this task, especially when fitting a surface to a new

data set.

Knot Selection for SPLINB

-------------------------

The following, somewhat heuristic, procedure works well in practice:

1. Use SELNOT to choose an initial set of knots from the data points.

This selects a specified number of data positions as knots by attempting

to equi-sample the independent spline variable space. While it is

theoretically possible to choose knots which are not at data positions,

this is not currently supported by SPLINB. In the limiting case where

every data position is a knot, then the actions of SPLINA and SPLINB are

identical. Choose an initial set of knots about 1/3 the size

of the data set. The number of knots required depends on the spatial

complexity of the data being fitted - more knots for more spatially

variable data. If the signal of the final result is within 10% of the

number of knots (the maximum possible signal), then the process should be

re-done starting with a larger initial knot set.

2. Run SPLINB, with the output list of data and fitted values, and

examine the largest residuals for data errors. Re-fit the surface if

necessary after data errors have been corrected, usually without the

output list of data and fitted values. Add to the knot index file the

indices of the largest 15-25 residuals which are not already knots and

re-fit the surface with these additional knots. Note that each knot

index refers to the sequence number in the original data file and not

the sequence number of the points actually selected. It is usual, but

not essential, to maintain the increasing order of the sequence numbers

in the knot index file. Residuals which are already associated with a

knot are identified by a minus sign, both in the output ranked residual

list, and in the list of data and fitted values.

3. Repeat the procedure of adding to the knot list the indices of the

largest 15-25 residuals and re-fitting the surface until the solution

stabilises or the variance estimates output by the program are in

approximate agreement with a priori estimates. This should normally

only be done two or three times since there is a risk of overfitting the

data if it is done too many times.

Interpretation of Output Statistics

-----------------------------------

The output statistics are best interpreted in relation to the partial

spline model for n observed data values zi given by

zi = f(xi) + b'yi + ei (i=1,...,n)

where each xi is a d-dimensional vector of independent spline variables,

f is an unknown smooth function of the xi, each yi is a p-dimensional

vector of parameters or covariates, b is a p-dimensional vector of

unknown coefficients of the yi and each ei is an independent, zero mean,

error term with variance = wi*sig2 where wi is termed the relative

variance (known) and sig2 is constant across all data points but often

unknown (Hutchinson,1991a). The model reduces, on the one hand, to an

ordinary thin plate spline model when there are no covariates (p=0) and,

on the other, to a simple multivariate linear regression model when

f(xi) is absent. The latter possiblility is not currently permitted by

the programs.

The signal degrees of freedom and the error degrees of freedom for each

surface should add up to n (the number of data points). The surface

fitting procedure is normally considered to have failed to find a

genuine optimum value of the smoothing parameter rho if either rho is

very small and the signal is the maximum possible (equal to the number

of knots) or rho is very large and the signal is the minimum possible (a

number which depends on the number of independent variables and the

order of the derivative). Both of these conditions are flagged by an

asterisk in the output.

The generalized cross validation (GCV), mean square residual (MSR) and

the error variance estimate (VAR) are given together with their square

roots (RTGCV, RTMSR, RTVAR) which are in the units of the data. The

mean square residual is weighted according to the relative variance

estimates wi as provided in the data file. For the GCV calculation

these relative variances are rescaled to have average value 1 in order

to facilitate comparisons of GCV values across different models. If the

relative variance estimates are actually estimates of the absolute value

of the error variance (so that sig2=1), then the variance estimate

provided by the procedure should be approximately 1. The goodness of

fit of the fitted model may be checked by comparing the residual sum of

squares (n*MSR where n is the number of data points) with the critical

points of a sig2^*chi-squared variable with df degrees of freedom, where

df = the error degrees of freedom and sig2^ is an a priori estimate of

sig2.

When an estimate of sig2 is available an alternative strategy is to

provide the corresponding standard deviation estimate (sqrt(sig2)) to

the program (instead of -1). The program then minimises an unbiased

estimate of the true mean square error instead of the GCV. This is not

normally recommended since it depends on having a reasonably accurate

estimate of sig2. It is generally preferable to minimise GCV, since

this appears to be more robust and does not depend on knowing sig2. The

a priori estimate of sig2 can then be used to check the goodness of fit

of the model as described above. Specifying the error variance may be

preferable when fitting surfaces to very small data sets (less than say

20-30 data points). Note that SPLINA should be used for small data

sets.

The procedure also outputs the coefficients of any covariates (standard

errors of these coefficients are not yet available) as well as an

estimate of the true mean square error of the smoothed data values.

评论收藏

内容反馈

xinjane

2012-04-24

这个软件是采用薄片多变量光滑样条技术所建立的降水、温度、太阳辐射以及生物气象要素在不同时间尺度的空间分布数据信息系统。对于气象学上空间插值来说具有较高的精度，而且其本身操作系统很小巧方便，很实用。国外用得比较多，我觉得国内也应该普及一下，只是可能资源不太好找吧。
suifengzai

2013-09-21

有没有详细说明啊
Kernel23

2013-05-28

薄板样条插值工具，很好用。
huosuanni

2012-05-15

没有详细的说明文件，上手不是很容易
飞天小猪

2011-11-20

很小巧的插值软件，DOS界面操作，效率高。。。适合理论学习