Help for SIMPLIFY

 PURPOSE:

SIMPLIFY is a VICAR applications program used to remove high frequency
information (noise) components from digital images.  It is usually used
to remove visually distracting detail from classified or stratified images
such as images derived from the operation of VICAR program FASTCLAS.  Such
images usually are composed of a number of regions, where each pixel in a
region has the same DN except possibly for some isolated noise pixels.
SIMPLIFY changes the DNs of these isolated noise pixels to match the DN of
the surrounding pixels and writes the resulting image to an output file.
 EXECUTION:

The DNs for the regions in the input image are assumed to lie in the range
from 1 to some upper limit specified by the user through the NCLASS parameter.
No assumptions are made on the DNs of noise pixels.  Pixels with a DN of 0
are considered to be unclassified (not associated with any region).

In order to determine which DNs to change, SIMPLIFY applies a test to each
pixel in the input image.  The test involves finding the most frequently
occurring DN in a square window centered around the pixel being tested.
The exact form of the test depends on the various parameters specified by the
user, but in order to provide an overview, the test for the case where neither
the REPLACE nor the VALUE option is used is described below.  Much of the
description also applies to when REPLACE or VALUE is used.  (See under
REPLACE and VALUE.)

For the sake of determining the most frequently occurring DN, only DNs from 1
to the value of the NCLASS parameter are considered.  In case of a tie in
determining the most frequent DN, the lowest of the DNs in the tie is used.
Since the window is centered around the pixel being tested, we will refer to
the pixel being tested as the center pixel and will refer to its DN as the
center DN. When the center DN is nonzero it is changed to the most frequent DN
if the most frequent DN occurs more times than the center DN and the number of
times the most frequent DN occurs is greater than or equal to the threshold
(THRESHLD) parameter.  (This insures that a nonzero DN is not changed if no DN
occurs a substantial number of times in the window or if the center DN occurs
just as many times as the most frequent DN.)  When the center DN is zero it is
changed to the most frequent DN if the number of times the most frequent DN
occurs is greater than or equal to the threshold parameter.  When a DN is
changed, the change is not included in the subsequent windows. 
TAE COMMAND LINE FORMAT
      The following command line formats show the major allowable forms:
      SIMPLIFY INP=a OUT=b SIZE=(sl,ss,nl,ns) optional parameters
      SIMPLIFY INP=a OUT=b  SL=sl SS=ss NL=nl NS=ns optional parameters
      SIMPLIFY a b (sl,ss,nl,ns) optional parameters
      SIMPLIFY a b optional parameters

       Here 'a' represents the input image file name, and
       'b' represents the output image file name.
EXAMPLE

      SIMPLIFY INP=A OUT=B NCLASS=9 THRESHLD=8 WINDOW=5 'REPLACE

      In this example a five by five window is used.  The most frequent DN
      must occur at least 8 out of 25 times in a window to be used to replace
      a center DN.  Unclassified pixels will be replaced where possible.


RESTRICTIONS
1. The input and output images must be byte data.
 OPERATION:

The operation of SIMPLIFY is similar to the operation of any box-filtering
routine with the exception that surface smoothing DOES NOT take place.  In a
pictorial sense, a window is passed over each pixel in the input image.  For
each pixel and its surrounding neighborhood, defined by the size of the window,
a histogram is accumulated to identify the most most frequently occurring DN
value.  When the center DN is nonzero it is changed to the most frequent DN 
if the most frequent DN occurs more times than the center DN and the number of
times the most frequent DN occurs is greater than or equal to the threshold 
parameter.  When the center DN is zero it is changed to the most frequent 
DN if the number of times the most frequent DN occurs is greater than or 
equal to the threshold parameter.  

In order to accommodate the edges of the image which normally could not be
simplified due to edge overlap of the window, the pixels outside the image 
are considered unclassified (DN=0).  This action eliminates the need for 
special edge processing feature.

The following window operations will demonstrate the operation of SIMPLIFY
on various windows.

Example 1:  specifying THRESHLD=4

                    5   5   5                   5   5   5 

            INPUT:  3   2   5          OUTPUT:  3   5   5

                    4   3   3                   4   3   3

Example 2:  specifying THRESHLD=4

                    5   5   5

            INPUT:  3   2   2          OUTPUT:  no change

                    2   2   2 

Example 3:  specifying THRESHLD=4 and REPLACE

                    0   0   0                   0   0   0

            INPUT:  0   0   0          OUTPUT:  0   2   0

                    2   2   1                   2   2   1


 WRITTEN BY:             Steve Pohorsky               1 Dec 1983

 COGNIZANT PROGRAMMER:   A. L. Zobrist               14 Sept. 1984

 REVISION:               2                           21 Feb 1982

PARAMETERS: