Help for NACMA

PURPOSE

   Glynn Hulley and Simon Hook deserves much credit for this
work.  It was their idea to apply the LANDSAT cloud detection
algorithm to ASTER and their MATLAB scripts were used as
the groundtruth for this program.

   This program creates a cloud mask from a given ASTER image.
The output is a binary image where clouds are marked as 1s and
nonclouds are marked as 0s.  ASTER band images are passed to
the program in the inp parameter and are ordered 1, 2, 3, 4, 5,
13, and 14.

   This program will create temporary working files needed for
storing reflectance values and downsampled reflectance values.
Also tempcomp, ndsi, gv (growing vegetation), sv (scenesing
vegatation), rs (reflective soil), are also store in temporary
files.

   If no output files are specified for the temporary data then
default names will be given.  Please see parameter list for the
default names.

   The user may need to grow the final cloud mask by a few
pixels.

EXECUTION

   The program consists of 2 pass cloud detection system.  The
general run of the program is:

   1. calculate reflectance from raw dn
   2. calculate brightness temperature from TIR raw dn
   3. downsample reflectance files
   4. perform pass1
   5. perform pass2

   For details of the pass1 and pass2 algorithms, please see
"Characterization of the Landsat-7 ETM+ Automated Cloud-Cover
Assessment (ACCA) Algorithm" by Richard R. Irish,
John L. Barker, Samuel N. Goward, and Terry Arvidson.  The
article can be found in "Photogrammetric Engineering & Remote
Sensing", Vol. 72, No. 10, October 2006, pp. 1179 - 1188.

EXAMPLES

   The only parameters that are necessary are ASTER bands
(1, 2, 3, 4, 5, 13, 14) as inp, metafile, and output mask
file as out.  Other files can be specified for debugging
purposes.

   nacma inp=(b1, b2, b3, b4, b5, b13, b14) meta=meta          out=cloudmask

RESTRICTIONS

Original Programmer: P. Kim, 18 Aug 2009

Revisions:
  2016-05-16 W. Bunch - Migrated to MIPL; replaced printf with zifmessage