| Description: |
This ImageJ PlugInFilter fits a straight line to a set of
data points in a ResultsTable or in a file. The slope and
intercept of successive fits are saved as
lines in a text window. Also saved is the chi square value and a
goodness-of-fit parameter, 1-P, where P is the usual P-value.
An estimate of the Standard Deviation, SD, due to noise in the data
(True SD) can be entered (see
Numerical Recipes 15.2). If the "True SD" is not entered, the
Chi-sq and goodness-of-fit values are affected.
There are four possible methods fitting methods. Initially, this
plugin (1) searches for the N points with the largest slope.
Optionally, (2) it then adds points to the start and end so long as a
figure-of-merit (FOM) increases. On subsequent fitting steps, it
either (3) fits between a given start and end range or (4) searches for
the set of N or more points within that range which have the largest
figure-of-merit. Use weights of 0.0 for the second and third term
in the FOM in order to search for the N points with the greatest slope.
Install with "about" in the argument and an about message is printed.
FitSliceTimes.txt allows different time durations for different slices
in the stack. It is looked for first in the directory which
contains the source file (if any). If it is not found there, then
the plugins/NucMed/ directory is search. FitSliceTimes.txt lists
groups of slices. The Nslices column gives the number of slices
in a group and the Time column gives the time per slice. For
example, a file with the following contents describes 120 slices
collected at 1 second per slice followed by 20 slices collected at 15
seconds per slice.
|
Nslices
|
Time
|
1
|
120
|
1
|
2
|
20
|
15
|
FitSliceTimes.txt can be edited with any text editor. The fields
are separated by the <tab> character.
|