Differences between revisions 28 and 41 (spanning 13 versions)
Revision 28 as of 2016-02-25 14:45:44
Size: 5379
Editor: JohnWeirich
Comment:
Revision 41 as of 2016-02-26 08:37:26
Size: 3738
Editor: JohnWeirich
Comment:
Deletions are marked like this. Additions are marked like this.
Line 79: Line 79:
{{{
cd test
rm LMRKLIST1.TXT
cat support/TILE20.in
ls -l /usr/local/bin/bigmap
/usr/local/bin/bigmap < support/TILE20.in | tee log-TILE20
echo "TILE20" | showmap
convert TILE20.pgm TILE20.jpg
cp TILE20.jpg ~/send/
Use the following tile block with 20cm GSD.
Line 89: Line 81:
rm MAPFILES/XXXXXX.MAP
ln -s TILE20.MAP MAPFILES/XXXXXX.MAP

rm make_scriptT.in
ln -s lsupport/tile20_make_scriptT.in make_scriptT.in
cat make_scriptT.in
make_scriptT
ls -l

cat INIT_LITHOS.TXT
lithos -v
nohup sh run_script.b &

#since this will take awhile
exit
#copy typescript file somewhere...

#when finished tiling
script

cd test
date
pwd
ls -l
[[F3H-Tile]]
Line 115: Line 84:
find_nofitT == 4. Iterate ==
Line 117: Line 86:
#fix landmarks
#after fixing, continue on
[[F3H-Iterate]]
Line 120: Line 88:
map_coverage
XXXXXX
0 .0002
convert coverage_m.pgm coverage_m-TILE20-afterTiling.jpg
cp coverage_m-TILE20-afterTiling.jpg ~/send/
== 5. Tile at 10cm ==
[[F3H-Tile]]
Line 126: Line 91:
mkdir log/TILE20
find_nofitT | tee log/TILE20/log-fit
sh support/logEvalP.sh
cat logResults.txt
cat lowImg.txt
cat redo.txt
cp *INN *OOT run_script*.b log/TILE20
== 6. Iterate ==
[[F3H-Iterate]]
Line 134: Line 94:
# Clean directory
sh rem_script.b
rm TESTFILES/*
rm TESTFILES1/*
== 7. Clean landmarks ==
Execute [[Block 7 - Clean]]
Line 139: Line 97:
#copy typescript somewhere...
}}}
== 8. Geometry ==
Line 142: Line 99:
== Iterate ==
{{{
script
== 9. Tile at 5cm ==
Line 146: Line 101:
cd test
date
pwd
ls -l
== 10. Iterate ==
Line 151: Line 103:
cat support/TAG100.in
ls -l /usr/local/bin/bigmap
/usr/local/bin/bigmap < support/TAG100.in
cp USED_MAPS.TXT LIST.TXT
duplicates
== 11. Clean landmarks ==
Line 157: Line 105:
cat make_script.in
ls -l /usr/local/bin/MAKE_LMRKLISTX
MAKE_LMRKLISTX
ln -s scripts/make_scriptP.seed make_scriptP.seed
cat make_scriptP.seed
make_scriptP | tee run.sh
ls -l

# Invoke batch iteration
cat INIT_LITHOS.TXT
lithosP -v
nohup sh run.sh &

#since this will take awhile
exit
#copy typescript file somewhere...

#when finished iterating

#clean landmarks

script

cd test
date
pwd
ls -l

# Check results
find_nofitP | tee log-fitP
sh support/logEvalP.sh
cat logResults.txt
cat lowImg.txt
cat redo.txt

# Fix uncorrelated images
# Manual correlation as required

# Clean directory
sh rem_script.b
rm TESTFILES/*
rm TESTFILES1/*

#copy typescript somewhere
}}}
== 12. Run Stats ==

Test F3H - Latitude

Goal: Physical characteristics of the object. The maximum variation in sub-solar position is a function of the latitude of the TAG site. The higher in latitude the TAG site is, the lower quality the variations in Sun vector. Thus, we can get a better idea about how big this error is as we move to the polar regions.

Test Set Up

Data: Detailed Survey.

Start: 75cm Preliminary Survey Shape Model.

Build: Build TAG Site 1 for each Sub-Test, but using different images than standard. Because the shape model has different topography, we would not be able to build a TAG Site at each latitude and compare results -- RMS is strongly effected by variations in topography. As such, we will adjust the position of the Sun so that the images match the observing/illumination conditions that DS would have if TAG Site 1 was at different latitudes.

Sub-Test F3H-1 to F3H-9

For each Sub-Test, we start with the Preliminary Survey Shape Model. We build the topography using the Detailed Survey Images that are available at each latitude. Images will need to be generated with Bennu with the Sun position that cannot really exist. Either Bennu can be inclined by the needed number of degrees with the pole and date matching the summer solstist, or the SPOC has tools to generate images and SUMFILES for specific illumination geometries. The SPOC has conducted similar variations in illumination parameters so this is well understood. For this test, the SPOC would use Imager_MG to generate the images, which generated almost all of the images needed for Test F2-Calibration. The Test Engineer has compared the output of Freespace and Imager_MG and has determined that the variations between the two systems are not significant.

Details for Sub-Test F3H-1 to F3H-9:

Sub-Test

Latitude

Number of Images

Emission Range

Incidence Range

F3H-1

TBD

TBD

TBD

F3H-2

-8°

TBD

TBD

TBD

F3H-3

-20°

TBD

TBD

TBD

F3H-4

-30°

TBD

TBD

TBD

F3H-5

-40°

TBD

TBD

TBD

F3H-6

-50°

TBD

TBD

TBD

F3H-7

-60°

TBD

TBD

TBD

F3H-8

-70°

TBD

TBD

TBD

F3H-9

-80°

TBD

TBD

TBD

Evaluate:

  • Relative Accuracy. Create truth models for each of the latitude TAG sties. Then compute the RMS deviation between the truth and the model.

Examples:

F3H-ex1.png F3H-ex2.png F3H-ex3.png F3H-ex4.png

Directory Setup

Run from the F3H-Lat directory

0. Start typescript

script
date
pwd
ls -l

1. Create Sub-Test directory

rsync --delete  -ha orig/ test/

2. Add Sub-Test images

  • /SPC_Test/F3H-Lat/base/IMAGEFILES/ should already have all the images. We will control the images used by SPC via PICTLIST.TXT

  • Should also set up a PICTLIST.TXT for each sub-test and put it in lsupport/

2.1 User Input

testID=F3H-X

cp registered_images/$testID/NOMINALS/* test/NOMINALS/
ls -l test/NOMINALS
cp registered_images/$testID/SUMFILES/* test/SUMFILES/
ls -l test/SUMFILES
rm test/PICTLIST.TXT
ln -s lsupport/PICTLIST_$testID.TXT test/PICTLIST.TXT
cat test/PICTLIST.TXT
mkdir test/log

3. Tile at 20cm

Use the following tile block with 20cm GSD.

F3H-Tile

4. Iterate

F3H-Iterate

5. Tile at 10cm

F3H-Tile

6. Iterate

F3H-Iterate

7. Clean landmarks

Execute Block 7 - Clean

8. Geometry

9. Tile at 5cm

10. Iterate

11. Clean landmarks

12. Run Stats

TestF3H (last edited 2016-03-16 08:34:58 by JohnWeirich)