As the amount of scientific data continues to grow, researchers need new tools to help them visualize complex data. Immersive data-visualisations are helpful, yet fail to provide tactile feedback and sensory feedback on spatial orientation, as provided from tangible objects.
The production of a tangible representation of a scientific data set is one step in a line of scientific thinking, leading from the physical world into scientific reasoning and back: The process starts with a physical observation, or from a data stream generated by an environmental sensor. This data stream is turned into a geo-referenced data set. This data is turned into a volume representation which is converted into command sequences for the printing device, leading to the creation of a 3D printout via additive manufacturing (“3D-printing”). As a last, but crucial step, this new object has to be documented and linked to the associated metadata, and curated in long term repositories to preserve its scientific meaning and context.
This presentation showcases a reference workflow to produce tangible 3D data-prints based on Free and Open Source Software (FOSS), using both GRASS GIS and Paraview. The workflow was successfully validated in various application scenarios using a RapMan printer to create 3D specimens of elevation models, geological underground models, ice penetrating radar soundings for planetology, and space time stacks for Tsunami model quality assessment.
Seismic Method Estimate velocity from seismic data.pptx
3D-printing with GRASS GIS – a work in progress in report FOSS4G 2014
1. Scientific 3D Printing with GRASS GIS
Introduction and Work in Progress Report
Dr. Peter Löwe
FOSS4G 2014
2014-09-12
2. 2
In a nutshell
• Consumer 3D printers are evolving very quickly
• 3D printing extends „flat“ 2D science communication.
• 3D pre-prints can already be generated with GRASS GIS.
• Dedicated GRASS support for 3D printing will soon simplify the
process.
• Research libraries take on Visual Analytics and 3D printing,
driving the standardisation of Metadata.
• This is work in progress: We barely got started.
3. 3
The Library Angle:
Open Source + Science = Open Science
Science advances only if knowledge is shared.
Accelerating the sharing of scientific knowledge
accelerates the advancement of science.
4. 4
The Who ?
German: Technische Informationsbibliothek (TIB)
• largest science and technology library globally
• over 9 Mio. items,
• 180 Mio. documents
• 125 km of shelving
• national library of Germany for
• engineering, technology, and the physical sciences.
• funded by the Federal Ministry of Education and Research and the
16 German states.
• the world's first Digital Object Identifier (DOI) registration
agency for research data sets (since 2005).
• operates in conjunction with the Leibniz University, Hannover.
5. 5
The future:
Data-driven Libraries
„Libraries are changing from repositories
for journals and books to
engaged community centers offering
new services, shaping
innovative research.“
Libraries offer places and services for discovery.
The path to a relevant, 21st-century library: “serendipitous discovery.”
Christopher Erdmann, 2014
John G Wolbach Library at the Harvard-Smithsonian Center for Astrophysics
https://www.insidehighered.com/blogs/higher-ed-beta/data-driven-library-future
6. 6
„While scientists focus on the final frontier,
(data-driven libraries) will work on designing
a different kind of space
full of physical and virtual tools
that
capture imagination and
enable researchers to explore it.“
Christopher Erdmann, 2014
John G Wolbach Library at the Harvard-Smithsonian Center for Astrophysics
https://www.insidehighered.com/blogs/higher-ed-beta/data-driven-library-future
http://thrilling-tales.webomator.com
The future:
Data-driven Libraries
7. • Content based Retrieval
• Science 2.0 and Open Science
7 7
Applied research topics at TIB
•Visual Analytics
•Ontologies
8. 8
Visual Analytics: Querying and communicating Data
The challenge:
• Communicating the meaning of scientic data
• Haptic/Tactile Visualization ?
The need:
A tangible representation of scientific results.
?
1492 Today Future
9. 9
3D printing for science communication:
The larger picture
Target group
Transformation / Reduction of content
Scientific
Data
3D
Print
Scientist
Science
Communication
10. 10
3D printing for science communication:
The larger picture
Target group
Technical Printing Process
Metadata Management
Scientific
Data
3D
Print
Scientist
Science
Communication
Librarian / Data-Scientist
11. 11
3D printing for science communication:
The value of metadata
3D
Print
Back-linking by metadata
Inquiries
Technical Printing Process
Metadata Management
Scientific
Data
Librarian / Data-Scientist
13. 13
NMC Horizon Report 2014 Library Edition
on scientific libraries.
Trends, technologies and challenges over the next five years:
• Focus on research data and new forms of multidisciplinary research.
• Rise of alternative search technologies
• Need for radical change owing to technological and social upheaval.
• Released in May 2014
• http://www.nmc.org/publications/2014-horizon-report-library
• Created by
• New Media Consortium (NCM)
• ETH-Bibliothek Zürich,
• University of Applied Sciences HTW Chur
• German National Library of Science and Technology
15. 15
Top-Down :
DURAARK: DURable ARchitectural Knowledge
• European Commission-FP7-Project (2013 -
2016) focusing on methods and tools for
long- term preservation of 3D data sources
of architectural knowledge:
• Goal:
• Enrich Building Information Models with “as
built” information from scans
• Semantically enrich building models with
additional data sets
• Preserve 3D models for future reuse
16. 3D Data
(Geological
models,
soil
penetrating
radar)
Tsunami
Simulations
16
Bottom-up: GRASS GIS-based 3D-printing
Osaka City University
FabLab
Potsdam Materials
GIS modularisation
GIS workflow
development
Initial GIS-driven
experiments
Stakeholders
Printing
Process
Expertise
INAF
Astrophysics Institute,
Rome
Hardware
2D
Data
(Elevation
Models)
Data Sources
Software
Workflows and Services
TIB Hannover
GFZ Potsdam
Multiple linked learning processes
18. 18
„The future is here“ (again)
The potential of „3D printing“ as
featured in the news:
19. 19
„The future is here“ (again)
The potential of „3D printing“ as
featured in the news:
• Guns !
20. 20
„The future is here“ (again)
The potential of „3D printing“ as
featured in the news:
• Guns !
• Human body parts !
21. 21
„The future is here“ (again)
The potential of „3D printing“ as
featured in the news:
• Guns !
• Human body parts !
• Clothes !
22. 22
„The future is here“ (again)
The potential of „3D printing“ as
featured in the news:
• Guns !
• Human body parts !
• Clothes !
• Candy !
23. 23
„The future is here“ (again)
The potential of „3D printing“ as
featured in the news:
• Guns !
• Human body parts !
• Clothes !
• Candy !
• Space Exploration !
24. 24
3D Printing, the Gartner hype cycle, and science
http://surveys.peerproduction.net/wp-content/uploads/2012/11/GoogleTrendsGartnerHypeCycle.png
3D Printing
2014
25. 25
3D Printing, the Gartner hype cycle, and science
http://surveys.peerproduction.net/wp-content/uploads/2012/11/GoogleTrendsGartnerHypeCycle.png
3D Printing
2014
• Handpieces for science
communication
• among scientists
• towards the general
public
• Showpieces for exhibitions
• <your application goes here>
Reality-check
26. 26
Printing Process Overview
Data Conversion Model Export
Pre-print
3D print
GIS Domain
27. 27
Printing Process Overview
Data Conversion Model Export
Pre-print
3D print
Printing
domain
GIS Domain
30. 30
RapMan 3.2: Reality check
Marcel
Ludwig
Resident 3D
printing expert
at GFZ Potsdam
31. 31
RapMan 3.2: Reality check
Marcel
Ludwig
Resident 3D
printing expert
at GFZ Potsdam
32. 32
RapMan 3.2: Reality check
Raw
Material
Control
Unit
Print head,
cooling fan
Print in
progress
Marcel
Ludwig
Resident 3D
printing expert
at GFZ Potsdam
33. 33
Close-Up: Actual printing
Print
head
Internal
support
structure
External
support
structure
3D printing
in progress
43. 43
Complex data sets:
Tsunami propagation space-time-cubes
• Space Time Cube
(STC) of tsunami
wave propagation.
• Complex wave
propagation in time
and space.
• Allows visual model
quality assessment.
• Produced by GFZ
Potsdam
• On permanent
display at the Osaka
City University (2014)
3D Print
Tsunami
Model
Time
Lon/Lat
Tohoku
Shoreline
Pacific
45. 45
GRASS
GIS
Technical overview
Data
png/pdf
3D preprint:
vtk,VRML.etc.
3D preprints - just a gdal/ogr extension ?
2D Map-Printout
3D Volume-Printout
46. needs buffering
46
Fault 2
Structural integrity
GRASS
GIS
Technical overview
Data
png/pdf
Fault 1
3D preprint:
vtk, etc.
Geologic
stratum
Thematic generalisation
3D preprints - just a gdal/ogr extension ?
Tricky part,
Thematic generalisation and print consistency is required
47. 47
Technical overview:
Current situation
Data
2D/3D Data 3D 3D Triangulation
v.in.ogr
r.in.gdal
r.to.rast3 r3.mapcalc r3.out.vtk
new modules new modules new modules
Ingest
Volume
generation
Volume
processing
Export
t.xxx
48. 3D
Print
3D Printing
48
Capa-bilities
Raster
Wanted:
GRASS GIS 3D print workflow trailblazers
Volume generalisation with r3.x and t.x-modules
requires currently these skills:
• Science Interpreter/Communicator: „What
message to convey ?“
• Technical/Software:
• invent new workflows,
• script these,
• document them
• Admin/Pioneer: be able to install patches for
GRASS7, help improve code maturity
Volumes
Vector Time
Time
Volumes
Raster / Vector
52. 52
Use of USGS Digital Elevation Models
GRASS Clippings: 1992
• http://grass.osgeo.org/uploads/grass/history_docs/grassclip6_2_92.pdf
http://ned.usgs.gov/historic.html
53. 53
But how to make a before-after print
– in GRASS GIS ?
Pre-Eruption Post-Eruption Ejected material
55. 55
Next Step: Printout on a RapMan Printer
32 hours
3km^2
Z-scaling: 3*
56. 56
The road ahead
Workflow integration
• File formats (VRML, x3d, obj)
• Size of printout ?
• Coloring schemes ?
• Material selection ?
• this is currently left to the „printmaster“
Provenance, Metadata and Archiving:
• Not to be left to the scientists
• Leave it to the Libraries:
• „Handling metadata and indexes for over 5 millenia“
57. 57
Summary
• 3D printing extends „flat“ 2D science communication.
• 3D pre-prints can already be generated with GRASS GIS.
• Dedicated GRASS support for 3D printing will soon simplify the
process.
• Research libraries take on Visual Analytics and 3D printing,
driving the standardisation of Metadata.
• This is work in progress: We barely got started.
• .
58. 58
Thanks for listening
Have a great FOSS4G 2014 !
Contact: peter.loewe@tib.uni-hannover.de