file5.jp2 is not decoded correctly

[GoogleCodeExporter]

As per test ETS-JP2-file5.jp2-compare2ref file5.jp2 does not seems to be 
handled right.

Original issue reported on code.google.com by mathieu.malaterre on 12 Mar 2014 at 10:46

[GoogleCodeExporter]

Comparing the diff (compare from imagemagick) indicate that the image is a bit 
darker than kakadu.

Original comment by mathieu.malaterre on 12 Mar 2014 at 10:47

[GoogleCodeExporter]

Using kdu_transcode one can check that decompression is going perfectly fine. 
Thus issue lie within JP2 header (colr box)

Original comment by mathieu.malaterre on 12 Mar 2014 at 11:05

[GoogleCodeExporter]

the test suite report:

ERROR: MSE (2366.403376) or PEAK (101.000000) values produced by the decoded 
file are greater than the allowable error (respectively 1.000000 and 4.000000) 

However if one run the test suite with a file5.tif generated from kakadu one 
get:

ERROR: MSE (302.276098) or PEAK (50.000000) values produced by the decoded file 
are greater than the allowable error (respectively 1.000000 and 4.000000) 

Original comment by mathieu.malaterre on 12 Mar 2014 at 11:16

[GoogleCodeExporter]

I would be tempted to classify file7 test suite failure as similar as file5 
(colr box issues)

Original comment by mathieu.malaterre on 12 Mar 2014 at 11:28

• Changed title: file5.jp2 is not decoded correctly

[GoogleCodeExporter]

Using r2701, I am now able to retrieve the ICC profile as external file.

Original comment by mathieu.malaterre on 12 Mar 2014 at 1:56

[GoogleCodeExporter]

Steps:

$ kdu_transcode -i data/input/conformance/file5.jp2 -o file5.j2k
$ opj_decompress "-i" file5.j2k -o file5.tif
$ tificc -idebug.icm file5.tif file5.icc.tif
$ bin/compare_images "-b"  "data/baseline/conformance/jp2_5.tif" "-t" 
file5.icc.tif "-n" "3" "-p" "4:4:4" "-m" "1:1:1" "-s" "b_t_"
******Parameters********* 
 base_filename = data/baseline/conformance/jp2_5.tif
 test_filename = file5.icc.tif
 nb of Components = 3
 Non regression test = 0
 separator Base = _
 separator Test = _
 MSE values = [ 1.000000  1.000000  1.000000 ]
 PEAK values = [ 4.000000  4.000000  4.000000 ]
 Non-regression test = 0
 NbFilename to generate from base filename = 3
 NbFilename to generate from test filename = 3
************************* 
Step 1 -> Header comparison
Step 2 -> measurement comparison
<DartMeasurement name="PEAK_0" type="numeric/double"> 1.000000 
</DartMeasurement> 
<DartMeasurement name="MSE_0" type="numeric/double"> 0.085576 
</DartMeasurement> 
<DartMeasurement name="PEAK_1" type="numeric/double"> 1.000000 
</DartMeasurement> 
<DartMeasurement name="MSE_1" type="numeric/double"> 0.173566 
</DartMeasurement> 
<DartMeasurement name="PEAK_2" type="numeric/double"> 1.000000 
</DartMeasurement> 
<DartMeasurement name="MSE_2" type="numeric/double"> 0.204882 
</DartMeasurement> 
---- TEST SUCCEED ----

Original comment by mathieu.malaterre on 12 Mar 2014 at 1:58

[GoogleCodeExporter]

This issue was updated by revision r2702.

Original comment by mathieu.malaterre on 12 Mar 2014 at 2:17

[GoogleCodeExporter]

This issue was closed by revision r2703.

Original comment by mathieu.malaterre on 12 Mar 2014 at 2:18

• Changed state: Fixed

[GoogleCodeExporter]

KDU733_Demo_Apps_for_Linux-x86-64_140118
========================================

kdu_expand -i file5-ICC.jp2 -o file5-ICC-kdu.bmp

Kakadu Error:
Unable to find the compositing layer identified by your `-jpx_layer' argument.
Note that the first layer in the file has an index of 0.

kdu_transcode -i  file5.jp2 -o file5.j2k

Error in Kakadu File Format Support:
Unable to open compressed data file, "file5.j2k"!

KDU lottery :-(

winfried

Original comment by szukw...@arcor.de on 13 Mar 2014 at 4:28

[GoogleCodeExporter]

Library error. After hiding the old 'libkdu_v72R.so', I succeeded.

kdu_transcode -i file5.jp2 -o file5.j2k

Codestream #1/1
---------------
    Output contains 1 quality layers
    Transferred 309 code-blocks.

    Read 1 tile-part(s) from a total of 1 tile(s).
    Total bytes read = 585,880 = 11.919759 bits/pel.

    Wrote 1 tile-part(s) in a total of 1 tile(s).
    Total bytes written = 585,901 = 11.920186 bits/pel.
    All header (marker segment) bytes written = 121
    Layer bit-rates (possibly inexact if tiles are divided across tile-parts):
        11.920186

BUILD/bin/opj_decompress -i file5.j2k -o file5.tif
tificc -i/tmp/debug.icm file5.tif file5.icc.tif

BUILD/bin/compare_images "-b"  "jp2_5.tif" "-t" file5.icc.tif "-n" "3" "-p" 
"4:4:4" "-m" "1:1:1" "-s" "b_t_"

******Parameters*********
 base_filename = jp2_5.tif
 test_filename = file5.icc.tif
 nb of Components = 3
 Non regression test = 0
 separator Base = _
 separator Test = _
 MSE values = [ 1.000000  1.000000  1.000000 ]
 PEAK values = [ 4.000000  4.000000  4.000000 ]
 Non-regression test = 0
 NbFilename to generate from base filename = 3
 NbFilename to generate from test filename = 3
*************************
Step 1 -> Header comparison
Step 2 -> measurement comparison
<DartMeasurement name="PEAK_0" type="numeric/double"> 1.000000 
</DartMeasurement>
<DartMeasurement name="MSE_0" type="numeric/double"> 0.085576 
</DartMeasurement> 
<DartMeasurement name="PEAK_1" type="numeric/double"> 1.000000 
</DartMeasurement>
<DartMeasurement name="MSE_1" type="numeric/double"> 0.173566 
</DartMeasurement> 
<DartMeasurement name="PEAK_2" type="numeric/double"> 1.000000 
</DartMeasurement>
<DartMeasurement name="MSE_2" type="numeric/double"> 0.204882 
</DartMeasurement> 
---- TEST SUCCEED ----

winfried

Original comment by szukw...@arcor.de on 13 Mar 2014 at 5:52

[GoogleCodeExporter]

Original comment by antonin on 24 Mar 2014 at 3:57

• Added labels: Conformance