Printable Version of Topic

Click here to view this topic in its original format

Unmanned Spaceflight.com _ Image Processing Techniques _ Method used for annotating images.

Posted by: MouseOnMars Feb 12 2009, 04:07 PM

What method do Planetary Scientists use for annotating and commenting images ? Is there a formal method in use or is this done informally using a graphics package to add text and markers ? The reason I ask is that I am trying to develop such a method for my project Space Data Wiki. The advantage to having a formal method is that the annotations can be linked using the "Semantic Web" using RDF and OWL technologies (as I understand it at the moment). See "http://spacedatawiki.wiki.sourceforge.net/An+SDW+Web+2.0+Application#tocAn%20SDW%20Web%202.0%20Application11". Any information or suggestions will be much appreciated.

Posted by: djellison Feb 12 2009, 04:39 PM

Published annotations are so diverse in terms of style, quality, quantity and content that even if there were such a convention, it's never been stuck to.

Posted by: MouseOnMars Feb 13 2009, 11:12 AM

What are your sources ? Has there ever been a convention ? If there has been maybe it was applied in the wrong fashion. If there never has been, then I intend to try again. I appreciate that there are so many comment forms and types of annotation. But this is the purpose of having an organised approach to them and indeed an approach to investigating and categorising the increasing amounts of data coming back from space. What you may not be aware of is that I'm not the only one to notice this problem. This quote is from an introduction by the http://www.lpi.usra.edu/pcgmwg/ ...

"In this document the Planetary Cartography and Geologic
Mapping Working Group will assess the cartographic products, tools, and needs
for the planetary community in the decade 2005–2015. Of particular concern is
the vast amount of data anticipated from ongoing and planned missions."

Posted by: djellison Feb 13 2009, 11:16 AM

Data is not annotations. Yes - there are huge amounts of data anticipated - of course there are - and of course I'm aware of that. That has NOTHING to do with annotating images. Nothing.

There are PDS conventions for describing data - as I am sure you are aware - but, I say again, annotations are massively diverse things from scales to grids to labels to arrows to paths to sites to directions to layers to bedforms to routes to dunes etc etc - and all of these are done by JPL, Cornell, ESA, ESOC, SWRI, ASU, CICLOPS, HRSC, MPI, dozens of institutions annotating thousands of things in a million different way - there can be no convention on how to do that. Even two people doing exactly the same thing, do it differently.

http://marsrovers.jpl.nasa.gov/mission/tm-spirit/images/MERA_Sol1813_1_br2.jpg is different to http://marsrovers.jpl.nasa.gov/mission/tm-opportunity/images/MERB_Sol1687_2_br2.jpg

I would say you're looking for something that doesn't exist, and is impossible to establish. It's like trying to have one set of rules, for every sport in the world - from Football to F1. You need to better specify exactly what you mean by annotations before you can progress.

Posted by: imipak Feb 13 2009, 08:01 PM

MoM, probably not quite what you have in mind, but (in case you didn't already know) there are various standards for metadata in the digital image formats. Even JPEG includes EXIF metadata. For UMSF applications, see e.g. the Planetary Society overview at http://www.planetary.org/explore/topics/imaging/formats.html . The standard PDS format includes a metadata label; try http://pds.jpl.nasa.gov/documents/qs/labels.html , and also the examples linked from there.

Posted by: tedstryk Feb 13 2009, 09:01 PM

QUOTE (MouseOnMars @ Feb 13 2009, 12:12 PM)

What you may not be aware of is that I'm not the only one to notice this problem. This quote is from an introduction by the http://www.lpi.usra.edu/pcgmwg/ ...

"In this document the Planetary Cartography and Geologic
Mapping Working Group will assess the cartographic products, tools, and needs
for the planetary community in the decade 2005–2015. Of particular concern is
the vast amount of data anticipated from ongoing and planned missions."

What a great illustration to use in my logic class when I need an example of a non sequitur. There is no established convention on this stuff. And if one is ever developed, it won't be in a wiki, that's for sure.

Posted by: elakdawalla Feb 13 2009, 09:10 PM

Most scientists annotate their images in whatever image processing software they use, most often Photoshop. Annotations drawn on images have to be drawn to a scale that makes sense with their eventual display use -- very big for Powerpoint, smaller for print publication, even smaller for poster publication. So there's not a lot of value in a file format especially designed for storing image annotations. What some researchers are now doing is to use geographic information systems (GIS), so that information is stored in separate layers; that way the locations of the annotations are independent of whatever original image files were used for the mapping.

--Emily

Posted by: MouseOnMars Feb 16 2009, 06:44 AM

Thankyou. I will consider your replies.

QUOTE (elakdawalla @ Feb 13 2009, 09:10 PM)

What some researchers are now doing is to use geographic information systems (GIS), so that information is stored in separate layers; that way the locations of the annotations are independent of whatever original image files were used for the mapping.

--Emily

Which researchers ?
A bit demanding, aren't we? - Moderator