subimstat(1)                                                      subimstat(1)



NAME
       subimstat - Subtracts one average image from another with a t-test

SYNOPSIS
       subimstat

DESCRIPTION
       Subimstat subtracts one average image from another and uses standard
       deviation or variance images to find the statistical significance of
       the difference at each pixel, as evaluated by a t-statistic.  It then
       sets to zero all differences that are less significant than the speci-
       fied level of significance.  It can also output pixel values reflecting
       the level of significance rather than the difference.

       The average and standard deviation/variance images can be ones produced
       by Imavgstat, Clip, PEET, or other means.  When one starts the program,
       one designates a pair of A files (with average and S.D./variance
       images) and a pair of B files.  One can then subtract any section in B
       from any section in A; A and B may be the same pair of files.  Alterna-
       tively, a whole set of corresponding sections can be subtracted between
       the two files.

       The user is responsible for keeping track of how many samples were used
       in making each average, and informing this program of those numbers.
       The program needs these numbers to compute the t-statistic and evaluate
       its significance.

       Entries to the program:

       Average image file A
       Standard deviation or variance image file A
       Average image file B, or Return if same as file for A
       Standard deviation or variance image file B, or Return if same as file
          for A
       Output image file to store differences in

       0 if the files have standard deviations in them, or 1 if the files
          have variances

       0 to enter each pair of sections separately, with potentially different
       numbers of samples and significance levels, or 1 to enter a list of
       corresponding sections and treat them all the same.

       IF you entered 0, next enter:

         The number of differences to compute

         Then for each difference, enter:

            Section # in file A, section # in file B

            Number of samples making up those averages in A and B

            Significance level for setting differences to zero.  Enter a
               probability value such as 0.05, 0.01, etc.  Differences less
               significant than this value will be set to zero.  Enter a
               negative value to have significant pixels values set to
               the negative of the log of the probability, or to the positive
               log for negative differences.  For example, positive and
               negative differences with a P of 0.01 would be output as
               2 and -2, respectively.

       IF you entered 1 to treat a set of corresponding sections the same,
       next enter:

         The list of sections to subtract between the files, or / to do all
         sections

         The number of samples making up the averages in A and B

         The significance level for settings differences to zero, a positive
         or negative value as just described.


HISTORY
       Written by David Mastronarde,  1/27/90
       4/23/90 - have program look up criterion t value.
       4/12/95 - use local subroutines for t-value
       1/19/11 - allow list of corresponding sections

BUGS
       Email bug reports to mast at colorado dot edu.



BL3DEMC                              4.7.3                        subimstat(1)