What and Where: In 2019, at the 50th Lunar and Planetary Science Conference (LPSC), several participants gathered to discuss planetary technology-based topics. This meeting was held over lunch at the restaurant in the conference center and lasted about 1.5 hours. This is the 3rd or 4th such technology lunches held at LPSC. The unofficial tag-line for this meeting was “this is a no science zone”.
Why: LPSC is generally heavily science focused for talks and discussions, thus there are not many opportunities to discuss planetary technologies as a group. We note that abstracts and poster sessions are an excellent method to present technology topics, but again a larger group interaction is missing.
Who: The initial email invite went out from Trent Hare (Astrogeology) to essentially a random list of community members known to support various technology initiatives. Several others were able to attend from word of mouth. In total about 25 participants were able to attend. Facilities represented included NASA facilities (e.g. AMES), JPL (e.g. NAIF, PDS nodes, MIPL), USGS (Astrogeology), RPIF, APL, PSI, and many from U.S. and international universities (mission teams and students).
How: Due to the size of the group, the main topics discussed were guided. One of the main intents was to allow participants to see what others in the room are working on to hopefully spur on future collaborations. If you find a topic you are interested in, please reach out to share ideas, methods, software, and hopefully increase partnerships.
Topics:
Quick introduction for new Astrogeology community contacts:
· Robin Fergason (rfergason@usgs.gov) has taken over as lead technology manager for Astrogeology. Please contact her for any comments or issues related to any services Astrogeology supports.
· Jay Laura (jlaura@usgs.gov) took over as software lead at Astrogeology (including ISIS3). Jay introduced several community software initiatives at Astrogeology as we move to a more open source initiative for ISIS3 and other supported software tasks. https://github.com/USGS-Astrogeology. GitHub issues are being used for all projects. We want to hear about issues, feature requests, documentation improvements, etc.
o To facilitate this, Astrogeology software will also have oversite from a Technical Steering Committee (TSC) for several of their projects. For more see: https://github.com/USGS-Astrogeology/TSC.
o New discussion forums are available. https://astrodiscuss.usgs.gov. Sign-in using a github acnt.
o Please follow the Astrodiscuss to see updates and community Requests for Comments (RFCs). These will also available at the ISIS3 Wiki: https://github.com/USGS-Astrogeology/ISIS3/wiki.
o Note ISIS3 is also moving to a new versioning system. Thus, be prepared that version 4 is coming. This updated versioning better follows industry standards for software.
Image Matching and Change Detection
· The first question posed to the group was, “how many are interested in image matching techniques”. About 60% or more of participants raised their hands. This obviously has wide uses across several applications including automating image-to-image registration methods, image to “ground” registration (tying one image type to another image type or tying to a controlled base), terrain generation using stereo-pairs, facilitating change detection, etc.
· There are lots of papers and research in this topic. Simple searches provide many of the initial results below thus consider this a sampling of available resources (please send me more to add or include below in comments):
o Ames Stereo Pipeline: https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2018EA000409
o Improved SIFT: https://www.sciencedirect.com/science/article/pii/S0032063317303173
o Content based image matching: https://ti.arc.nasa.gov/m/pub-archive/1273h/1273%20(Deans).pdf
o Adaptive least squares: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2008JE003080
o Multi-instrument coregistration: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8077748
o FFTs (AFIDS): https://www.hou.usra.edu/meetings/lpsc2018/pdf/1178.pdf
o Optimizing change detection: https://iopscience.iop.org/article/10.1088/1742-6596/1036/1/012004/pdf
o Lunar impacts: https://meetingorganizer.copernicus.org/EPSC2018/EPSC2018-482.pdf
Machine Learning (and Feature Extraction)
· While not as popular as image matching, ML topics were also popular among participants. This is probably one of the fastest growing research areas for planetary (and Earth-based) domains. Like above, a simple search provides many papers and abstracts (please send me more to include or add in the comments below):
o Machine learning session at 2018 AGU: https://agu.confex.com/agu/fm18/prelim.cgi/Session/50013
o Novelty detection: https://hannah-rae.github.io/files/Kerner_et_al_2019_IAAI.pdf
o Interpretable discovery: http://www.wkiri.com/research/papers/wagstaff-interp-18.pdf
o Dune Fields: https://ieeexplore.ieee.org/document/5692810
o Lots of automatic crater detection papers!
NAIF and SPICE
· Chuck Acton stated the transition to a multi-thread SPICE library is proceeding well.
· Syncing SPICE kernels remains to be a large community issue. For example, if SPICE kernels are updated, how are they to be released? Sent back to NAIF? And how and who reviews them? As more groups are creating controlled mosaics, this issue will continue to grow.
· Cosmographia continues to be supported by NAIF. One idea stated by the group included making SPICE kernels more easily integrated into Cosmographia. For more: https://naif.jpl.nasa.gov/naif/cosmographia.html
· SpiceyPy, supported by Andrew Annex, has recently been updated to v2.2.0. https://github.com/AndrewAnnex/SpiceyPy.
· There is an upcoming (free) SPICE training (June 4-6, 2019). I believe all materials will be made available to all. To attend, please register: https://naif.jpl.nasa.gov/naif/WS2019_announcement.html
· While Astrogeology attempted to use Cosmographia for ISIS3’s control environment, it has been deprecated.
Planetary Web Mapping Services (WMS)
· Lunaserv is still being updated. It is the first WMS server to support Planetary codes. http://lunaserv.lroc.asu.edu/about.html
· There is a new branch of Mapserver by Jean-Christophe Malapert which also uses planetary codes. It is not yet pushed back into Mapserver baseline. For more information, contact Jean-Christophe on github.
Planetary Data System (PDS)
· There was a strong attendance of PDS personnel at LPSC.
· Thanks to Nick Estes’ idea, we have an initial method to archive GIS data in the PDS. We are using a variant of GeoCSV with a PDS4 label: https://giswiki.hsr.ch/GeoCSV. An early implementation is now in GDAL for testing: https://www.gdal.org/frmt_pds4.html
· Upcoming technical session for PDS personnel in April will cover many topics listed here and including PDS4, tools for PDS4, facilitating search, using Github, long-term roadmap, etc. Supporting the community and better advertising updates and direction for the PDS is also a main goal.
Upcoming Technical Workshops for the Planetary Domain
· The 4th Planetary Data Workshop (not free), June 18-20, Flagstaff, AZ: https://www.hou.usra.edu/meetings/planetdata2019/. It will have presentations, posters, tutorial demos, and discussions.
· The 2nd Planetary Mapping and Virtual Observatory Workshop, July 1-3, Domaine de St. Paul, Saint-Rémy-lès-Chevreuse, France. This a great “partner” workshop for the Planetary Data Workshop and covers many related technical topics. It will have presentations, discussion workshop, and tutorials. https://epn-vespa.github.io/mapping2019/
Recent Published Books (many technical chapters available)
· Planetary Remote Sensing and Mapping: https://www.taylorfrancis.com/books/9780429000515
· Planetary Cartography and GIS: https://www.springer.com/us/book/9783319628486
· Planetary Geology: https://www.springer.com/us/book/9783319651774
(Mario D'Amore here) I had to leave to early, this is a really nice thing Trent!
ReplyDeleteI was proposing the Machine learning session at 2018 AGU with other 3 co-proposer, we have a session this year at the EPSC-DPS meeting in France too > https://meetingorganizer.copernicus.org/EPSC-DPS2019/session/34102 . This is a technique-first session, were we are interested in the application and cross field mixing. Having science in it is nice too.