FY03 Warfighter Technical Council WTC IPR

1
A GEOSPATIALLY ENABLED, PC-BASED, SOFTWARE TO
FUSE AND INTERACTIVELY VISUALIZE LARGE 4D/5D DATA
SETS
Makai Ocean Engineering, Inc. Presentation
for National Oceanic and Atmospheric
Administration (NOAA) August 13, 2008 Presented
by Dr. Jose Andres McKay Davis Jose.Andres_at_maka
i.com McKay.Davis_at_makai.com
2
MAKAI OCEAN ENGINEERING

• Located on Oahu, Hawaii
• Areas of expertise
• GIS / Visualization Software
• Submarine CablePlanning Installation Software
• Ocean Renewable Energy

3
PROBLEM ADDRESSED EXISTING APPROACH

• Understanding 3D/4D large data sets
• Currently visualized using 2D layers
• True 3D/4D visualization currently performed on
  large supercomputer clusters

2D VIEWS
4
TECHNICAL CHALLENGE LIMITATIONS OF 3D/4D
VISUALIZATION (VOLUME RENDERING)

• Images terrain are associated with pixels (X,
  Y, color/value) i.e., 2D images
• 4D/5D is associated with a voxels (X, Y, Z, time,
  multiple values)
• Demands on memory to render 3D/4D data are two
  orders of magnitude greater than what is
  currently available with PC technology

5
NAVY APPLICATIONS (NAVOCEANO)
July 7
25 June
26 June
Regional Acoustic Assessment (Pd)
SNR transform to Pd (W.A. Albersheim)
MSRC
UNCLASSIFIED
6
ADVANCED CAPABILITIES (NAVOCEANO)
Probability of Detection (Pd) Map Varying in Time
Volume View for 5-25 July 06 Indicates Sea
Glider position every ¼ day
UNCLASSIFIED
7
WHAT ARE WE DEVELOPING?




8
MAKAIS GOALS
State of the Artalgorithms tovisualizeLARGE
amountsof volumetric datain a 3D/4D
interactivevirtualenvironment.

• Terrain volumetric visualization of large
  databases
• Fusing data from DIFFERENT sensors and databases
• One single display can render 200 MILLION data
  points
• Software allows for TIME VARYING data
• DISPLAYED ON PCs, not supercomputers

9
MAKAIS TECHNICAL APPROACH

• Makais visualization method is expected to be
  two orders of magnitude faster than conventional
  methods.This is achieved by using

• Semi-transparent colored iso-surfaces ? FAST
• New multi-threading capabilities of PC processors
• Tiling technology for terrain, imagery
  volumetric data
• Continuous Level of Detail (CLOD)
• Compression techniques to minimize file sizes

RESULT We can use PCs instead of supercomputers
10
Funding History

• CEROS (2006-2007)
• Algorithms development and basic PC-based
  software
• Proof-of-concept to retrieve data from remote
  locations
• ONR (2008)
• Enhanced, stand-alone PC based system
• Platform independent
• Spherical mapping
• Multiple formats (e.g. , netCDF)
• Multi-layers?

11
SAMPLE SCREENSHOTS
12
SAMPLE SCREENSHOT
13
Current Development (movie)
14
WHAT DATA CAN BE VISUALIZED?

• Terrain/bathymetry and imagery at very high
  resolution
• Sensors, moving assets, tracking targets
• Any variable (scalar or vector) that changes in
  X, Y, Z, time
• Very large datasets (GB to TB)
• Interactive visualization (gt 20 fps) on a desktop
  PC
• All data is geo-referenced
• Interface (I/O) with other GIS systems

15
CURRENT TECHNOLOGY

• Google Earth

• Web-based, interactive geospatial application
• Used for elevation and imagery data
• Cannot show bathymetry or time dependent
  volumetric data

ArcGIS Explorer (by ESRI)

• True GIS platform
• Used for viewing spatial data, creating maps, and
  perform spatial analysis
• Mainly for 2D data
• Not highly interactive, even for small 3D/4D data
  sets

16
NOAA INVOLVEMENT

• We would welcome the opportunity to work with
  NOAA to identify specific areas of application.
• Participating in our development, NOAA has the
  opportunity to shape our development to better
  fit your needs.
• Our development efforts would greatly benefit
  from your input and your technical and program
  support.

17

• END

18
(No Transcript)
19
GOALS for 2008-2009 A CEROS PROJECT

• Enhance data pre-processing software
• Develop Client-Server management software
• Develop advanced analysis tools (e.g., ESRI
  interfaces)
• Generate geo-referenced outputs for Command
  Control

20
Comparison with other NAVO Tools

• Makais 4D Visualization software will include
• Advanced terrain visualization ( Google)using
  tiling, CLOD, import multipleimages, GIS shape
  CAD files in thescene, track moving targets,
  tools tocreate labels, routes and polygons,
  additionof 3D objects, measure distances, load
  geo-referenced points from files, etc.
• 3D/4D visualization Show iso-surfaces,
  truevolume rendering and flow visualization in
  avariety of formats (NetCDF, ASCII,
  binary)using 32/64 bit processing, tiling, CLOD,
  andlatest GPU technology. Develop a variety of
  tools such as perform math operations with
  datasets, cutting slices and planes, measure
  distances between iso-surfaces, render data from
  sensors.
• Provide a high performance 3D graphics toolkit
  (i.e., OSG - used by application developers in
  the fields of visual simulation, games, virtual
  reality and modeling) to better control and
  enhance the scene.

21
Expansion Possibilities

• Include the ability to access data on-demand from
  a centralized or distributed server(s) network.
• Expand capabilities to visualize real-time data
  from sensors.
• Specific applications for NAVY (ASW, NSW, MWC,
  etc.)

22
Battlespace on Demand(The Three Tiers)
All Tiers to be developed simultaneously
Tier 3 the Decision Layer

• Options / Courses of Action
• Asset Allocation / Timing / Risk
• GOAL Provide Information via Web
  Services and scheduled, tailored
  push products

Tier 2 the Performance Layer
Tier 1 the (forecast)
Environment Layer
Initial and Boundary Conditions
UNCLASSIFIED
23
MAKAI APPROACH - 1

• New visualization method is expected to be two
  orders of magnitude faster than conventional
  methodsThis is achieved by using

• New Multi-threading capabilities of PCs
• Compression techniques to minimize file sizes

24
MAKAI APPROACH - 2

• Tiling technology for terrain, imagery
  volumetric data
• Semi-transparent colored iso-surfaces ? FAST

25
MAKAI APPROACH - 3

• Continuous Level of Detail (CLOD) once tile is
  loaded in memory
• RESULT We can use PCs instead of supercomputers