Inquiry for Metadata for Earth Observations Task (SIS-356)

[Aaron Artea <aa...@gmail.com>]

Hello,

I'm interested in helping out with the following task:

https://issues.apache.org/jira/browse/SIS-356

Please let me know how I can be of assistance.

Thanks!

Aaron

-- 

When in doubt, mumble; when in trouble, delegate; when in charge, ponder.

Re: Inquiry for Metadata for Earth Observations Task (SIS-356)

[Martin Desruisseaux <ma...@geomatys.com>]

Hello Aaron, and welcome!

Le 28/07/2017 à 15:15, Aaron Artea a écrit :

> I'm interested in helping out with the following task:
>
> https://issues.apache.org/jira/browse/SIS-356
>
> Please let me know how I can be of assistance.

Are you familiar with ISO 19115 metadata standard? If not, I suggest to
browse the NOAA wiki [1]. Do not paid attention to the XML
representations, only to the properties (e.g. "acquisitionInformation"
may contain an "instrument" property, etc.). Some pages have UML diagrams.

If you have some NetCDF data, you may try to execute the SIS command
line on them. The "Extraction ISO 19115 metadata" example at [2] shows how.

After you got a feeling of ISO 19115 (no need to know it in details),
next step would be to download the "NetCDF Earth Observation (EO)
Metadata Conventions" document (link in SIS-356). Then the hardest part
is to decide a mapping from the EO document to ISO 19115. Example: the
EO document specifies:

    phenomenon_time_begin_position: Acquisition start date time

In ISO 19115, the acquisition start date time can be stored there:

  * Start from Metadata [3]
  * Go to 'acquisitionInformation' node (getAcquisitionInformation() method)
  * Go to 'operation' node (in AcquisitionInformation)
  * Go to 'significantEvent' (in Operation)
  * In the Event node:
      o Set 'context' to "acquisition"
      o Set 'time' to the date

To make shorter, we just document the path like below:

    Metadata/acquisitionInformation/operation/significantEvent/time with Context.ACQUISITION.

So the first step would be to list such mapping in a document. We do not
need to map the whole EO document (it would be a lot of work). If we
could map just a few main properties, that would be a nice start.

After we documented some "EO to ISO 19115" mappings, the next step will
be to implement them. That would be the subject of another email.

    Regards,


[1] https://geo-ide.noaa.gov/wiki/index.php?title=Category:ISO_19115
[2] http://sis.apache.org/command-line.html
[3] http://www.geoapi.org/snapshot/javadoc/org/opengis/metadata/Metadata.html