Details
Description
Data product readers and writers shall be able to gather additional information (e.g. compression level) from the user (see interface javax.imageio.IIOParamController).
****** ADDITIONAL INFORMATION ******
Dear Norman,
thanks for your encouraging mail. We are eagerly waiting for the next issue of BEAM including those options.
If you have to decide between PNG and GIF please consider (also) PNG since it is a "freely" available format. For reading GIF instead a license is needed. Our educational software LEOWorks can read PNG but not GIF exacltly for this reason.
I understand that considering orignial MERIS data also consisting of 16bit, a reduction to 8bit as option for the geotiff export is possible. This would be great.
Regards
Juerg
--------------------------------------------------------------
Dr. Jürg Lichtenegger
Keltenstrasse 10
CH-8044 Zurich
Switzerland
Phone: 0041 (0)44 261 91 89
eMail: jlichtenegger@bluewin.ch
EDUSPACE Operational Team
http://www.eduspace.esa.int
---------------------------------------------------------------
----- Original Message ----- From: "Norman Fomferra" <norman.fomferra@brockmann-consult.de>
To: "juerg lichtenegger" <jlichtenegger@bluewin.ch>
Cc: "Leif Toudal" <ltp@oersted.dtu.dk>; "Peter Brøgger Sørensen" <brogger.p@inet.uni2.dk>; "carsten brockman" <carsten.brockmann@brockmann-consult.de>; "Marco Peters" <marco.peters@brockmann-consult.de>
Sent: Tuesday, May 09, 2006 8:04 AM
Subject: Re: Further BEAM requests
> Dear Juerg,
>
> We are indeed planning to parameterise the data product readers and writers in BEAM. Following your recommondation, the updated GeoTIFF product writer would then open a dialog and promting the user to specify the export options you've given below.
> Concerning your request, please note that MERIS L1b data originally is also 16bit. BEAM applies scaling factor + offset for convenience and stores the radiances as FLOAT. This is also because the GeoTIFF format does not honour scaling factors/offsets properties of a band. Secondly, exporting as 8bit data does not automatically involve JPEG compression. JPEG is a lossy compression which we do not recommend for measurement data. It is better to use a lossless run-length encoding algorithm as used for the PNG and GIF formats.
>
> Best regards
> Norman
>
>
> juerg lichtenegger schrieb:
>> Dear Norman,
>> as you can imagine BEAM is never good enough! And so having long discussions with the authors of our image data bank for the Galathea project for which wie mostly input ENVISAT data we come up with further wishes. Since the data will not only be used by scientific people but also be pupils in secondary schools they will have problems to deal with FLOAT and 16bit data.
>> But on the other hand geotifs are needed to locate the ship and landmarks.
>> Our request is to have an option when exporting to geotifs:
>> 1. export original data (for MERIS it is FLOAT, for ASAR it is 16 bit data)
>> 2. export as 8bit data (data reduction as you do when exporting to jpeg)
>>
>> I have checked the radiometry of the jpeg images. The compression to 8bit is very well done. For 16bit this operation is always very critical and ends up with either to many saturated pixels or with low contrast in the lower part of the data. But I can see from your method that saturation is acceptable.
>> Of course, calibration is lost. However it is still possible to recalibrate the image using some values taken from the original image. I think this is good enough for our purpose. In any case such calibration will be controlled by us. Scientists can still download original data (N1) or have them exported as FLOAT.
>>
>> What do you think, can this be done?
>> Regards
>> Juerg
>>
>> --------------------------------------------------------------
>> Dr. Jürg Lichtenegger
>> Keltenstrasse 10
>> CH-8044 Zurich
>> Switzerland
>> Phone: 0041 (0)44 261 91 89
>> eMail: jlichtenegger@bluewin.ch
>>
>> EDUSPACE Operational Team
>> http://www.eduspace.esa.int
>> ---------------------------------------------------------------
>>
>>
>
>
--
Internal Virus Database is out-of-date.
Checked by AVG Free Edition.
Version: 7.1.385 / Virus Database: 268.4.4/320 - Release Date: 20.04.2006
****** ADDITIONAL INFORMATION ******
Dear Norman,
thanks for your encouraging mail. We are eagerly waiting for the next issue of BEAM including those options.
If you have to decide between PNG and GIF please consider (also) PNG since it is a "freely" available format. For reading GIF instead a license is needed. Our educational software LEOWorks can read PNG but not GIF exacltly for this reason.
I understand that considering orignial MERIS data also consisting of 16bit, a reduction to 8bit as option for the geotiff export is possible. This would be great.
Regards
Juerg
--------------------------------------------------------------
Dr. Jürg Lichtenegger
Keltenstrasse 10
CH-8044 Zurich
Switzerland
Phone: 0041 (0)44 261 91 89
eMail: jlichtenegger@bluewin.ch
EDUSPACE Operational Team
http://www.eduspace.esa.int
---------------------------------------------------------------
----- Original Message ----- From: "Norman Fomferra" <norman.fomferra@brockmann-consult.de>
To: "juerg lichtenegger" <jlichtenegger@bluewin.ch>
Cc: "Leif Toudal" <ltp@oersted.dtu.dk>; "Peter Brøgger Sørensen" <brogger.p@inet.uni2.dk>; "carsten brockman" <carsten.brockmann@brockmann-consult.de>; "Marco Peters" <marco.peters@brockmann-consult.de>
Sent: Tuesday, May 09, 2006 8:04 AM
Subject: Re: Further BEAM requests
> Dear Juerg,
>
> We are indeed planning to parameterise the data product readers and writers in BEAM. Following your recommondation, the updated GeoTIFF product writer would then open a dialog and promting the user to specify the export options you've given below.
> Concerning your request, please note that MERIS L1b data originally is also 16bit. BEAM applies scaling factor + offset for convenience and stores the radiances as FLOAT. This is also because the GeoTIFF format does not honour scaling factors/offsets properties of a band. Secondly, exporting as 8bit data does not automatically involve JPEG compression. JPEG is a lossy compression which we do not recommend for measurement data. It is better to use a lossless run-length encoding algorithm as used for the PNG and GIF formats.
>
> Best regards
> Norman
>
>
> juerg lichtenegger schrieb:
>> Dear Norman,
>> as you can imagine BEAM is never good enough! And so having long discussions with the authors of our image data bank for the Galathea project for which wie mostly input ENVISAT data we come up with further wishes. Since the data will not only be used by scientific people but also be pupils in secondary schools they will have problems to deal with FLOAT and 16bit data.
>> But on the other hand geotifs are needed to locate the ship and landmarks.
>> Our request is to have an option when exporting to geotifs:
>> 1. export original data (for MERIS it is FLOAT, for ASAR it is 16 bit data)
>> 2. export as 8bit data (data reduction as you do when exporting to jpeg)
>>
>> I have checked the radiometry of the jpeg images. The compression to 8bit is very well done. For 16bit this operation is always very critical and ends up with either to many saturated pixels or with low contrast in the lower part of the data. But I can see from your method that saturation is acceptable.
>> Of course, calibration is lost. However it is still possible to recalibrate the image using some values taken from the original image. I think this is good enough for our purpose. In any case such calibration will be controlled by us. Scientists can still download original data (N1) or have them exported as FLOAT.
>>
>> What do you think, can this be done?
>> Regards
>> Juerg
>>
>> --------------------------------------------------------------
>> Dr. Jürg Lichtenegger
>> Keltenstrasse 10
>> CH-8044 Zurich
>> Switzerland
>> Phone: 0041 (0)44 261 91 89
>> eMail: jlichtenegger@bluewin.ch
>>
>> EDUSPACE Operational Team
>> http://www.eduspace.esa.int
>> ---------------------------------------------------------------
>>
>>
>
>
--
Internal Virus Database is out-of-date.
Checked by AVG Free Edition.
Version: 7.1.385 / Virus Database: 268.4.4/320 - Release Date: 20.04.2006
Issue Links
Activity
Olaf Kock
made changes -
| Field | Original Value | New Value |
|---|---|---|
| issue.field.mantisimportkey | 722 | 11338 |
Norman Fomferra
made changes -
| Affects Version/s | 3.6 and earlier [ 10067 ] | |
| Mantis ID | 722 | |
| Fix Version/s | 4.0.0 M2 [ 10084 ] | |
| Remaining Estimate | 1 week, 4 days [ 259200 ] | |
| Original Estimate | 1 week, 4 days [ 259200 ] |
Norman Fomferra
made changes -
| Status | In Progress [ 3 ] | Open [ 1 ] |
Norman Fomferra
made changes -
| Fix Version/s | 4.0.0 M3 [ 10085 ] | |
| Fix Version/s | 4.0.0 M2 [ 10084 ] |
Marco Peters
made changes -
| Fix Version/s | 4.0.0 [ 10086 ] | |
| Fix Version/s | 4.0.0 M3 [ 10085 ] |
Marco Peters
made changes -
| Fix Version/s | 4.0.0 [ 10086 ] | |
| Fix Version/s | 4.1 [ 10132 ] |
Marco Peters
made changes -
| Fix Version/s | 4.1 [ 10132 ] |
Norman Fomferra
made changes -
| Fix Version/s | 4.1 [ 10132 ] |
Norman Fomferra
made changes -
| Fix Version/s | 4.1 [ 10132 ] | |
| Fix Version/s | 4.2 [ 10200 ] |
Norman Fomferra
made changes -
| Fix Version/s | 4.2 [ 10200 ] | |
| Fix Version/s | 4.5 [ 10240 ] |
Norman Fomferra
made changes -
| Fix Version/s | 4.5 [ 10240 ] | |
| Fix Version/s | 4.5.1 [ 10250 ] |
Norman Fomferra
made changes -
| Fix Version/s | 4.5.1 [ 10250 ] | |
| Fix Version/s | 4.6-SNAPSHOT [ 10260 ] |
Norman Fomferra
made changes -
| Fix Version/s | 4.6-SNAPSHOT [ 10260 ] |
Marco Peters
made changes -
Marco Peters
made changes -
Norman Fomferra
made changes -
| Fix Version/s | 5.0 [ 10431 ] |