An Extensible 3D Visualization Architecture for Distribuited Data Processing

1
An Extensible 3D Visualization Architecture for
Distribuited Data Processing

• Luigi Mazzucchelli, Ph. D.
• NEXT S.p.A.
• Luigi.mazzucchelli_at_next.it

Matteo Pace Department of Computer Science Univ.
La Sapienza ROME
2
Summary

• Introduction
• ASI-3D project
• Architecture benefits
• Converting data
• Layers
• Area representation
• XML in ASI 3D
• Mass context and Asi-3D

3
Introduction

• Satellite position based applications and
  location based information services, especially
  earth observation (EO), have had a great
  influence on the development of new applications
  in various fields.
• Today the transformation from the Satellite
  products to the final user document is performed
  by a number of specialized processes.
• Every process may use proprietary format
  so it seems very difficult for a final user to
  have a general information area that cover all
  the aspects.
• ASI-3D project resulted in an open system
  infrastructure based on a scalable component
  based architecture, capable to distributely
  process large data-sets coming from heterogeneous
  sources and to provide 3D scenarios in VRML/XML
  formats

4
The project ASI-3D
The presented system, called ASI-3D, is a
research and development project realized by the
NEXT s.p.a. in collaboration with ASI (Italian
Space Agency). It aims to study stereoscopic
visualization techniques as rappresentation of
georeferenced data in 3D fashion. It can
exhibit simultaneously various kind of data
without affecting the performance of the user
navigation in the virtual scenario by the
adoption of an open infrastucture able to manage
the addition of brand new components dedicated to
the elaboration of additional data or graphic
layers. This system uses auto-stereoscopic
displays and 3D mouse as visualisation and
navigation devices.
5
ASI-3D Context
3D visualization results from the integration of
heterogeneous data, structured into layers, and
produced, in turn, by heavy elaborations and
complex data-retrieval where throughput and
elaboration limitations drive to a distributed
multiprocessor architecture. Presenting data in
a 3D mode is the ability to show, analyse,
manipulate and interact with data in a
three-dimensional virtual space. In this way the
user is able to interact with the data in
(near)real-time mode, acting consequently to its
interaction with the scenario.
6
ASI-3D system

• ASI 3D is a distribuited and extensible system,
  taking many different types of input data in
  parallel in which representations may include
• Orographic and meteorological
  
  layers (2,5D or 3D
  models)
• isobaric and wind fields layers
• A radar navigation layer (used to display
  airplane traces)
• A balloon navigation layer (used to display
  weather balloon traces)

Scenario VRML 3D
7
ASI-3D components
Data sets
Data sets
CORBA
8
Architecture benefits (1/3)

• High scalability due to parallel computation on
  suitable functionalities (e.g. sub-domain
  computations).
• Extensibility for the representation of new kind
  of data information.
• Open Component Distributed architecture for the
  integration of COTS such as VRML clients or data
  modelling tools such as Matlab.
• Based on standard integration bus CORBA.
• Architectural independence from
  Telecommunication System using CORBA.

9
Architecture benefits (2/3)

• Pattern Oriented architecture that gives
  advantages in terms of robustness.
• Push model Data Transfer in near-real time.
• Dynamic Data Push in 3D environment
  representation during navigation in seamless
  mode.
• Light payload (small size file) is possible
  using VRML language to represent 3D scenarios.
  
• XML representations are also possible.
• A central controller (Director) controls and
  manages the data/control flow among the
  different components.

10
Architecture benefits(3/3)

• Every addition of data translators is performed
  by a subscription to Mediator patterns, that
  decrease the coupling degree between the
  components in the system allowing an easier
  substitution or replacement of the components
  with new and better ones.
• A communication protocol through an IDL interface
  has been also used to distribute events and
  messages (Design Pattern Publisher/Subscribers).
• Any component can be either added or removed at
  run-time, enabling a dynamic and extendible
  environment.
• COTS components can be seamlessly integrated in
  the architecture using a particular pattern which
  encapsulates low-level functions and data
  structures.

11
Converting data
Data components take the responsibility of
converting data to the visualization format,
currently (Geo)VRML (Georeferenced Virtual
Reality Modelling Language). Whenever updated
data (in near real time mode) are available, they
are pushed to the visualization component with
the mediation of the coordinator component. The
virtual scenario is thus incrementally depicted
and is navigable in 3D modality by the user with
the use of a 3D mouse that allows a more deep
understanding of data relations and seamless
navigation
12
Scenario representation
The scenario representation is actually in
GeoVRML (the geographic extension of the VRML)
format. The VRML is a modelling language and it
is, therefore, complementary to a markup language
since it defines a quite different set of
primitive and importantly three-dimensional
objects which can be used to express complex
models in a visual manner.
13
XML in ASI 3D
SAT data
The system finally transforms the satellite data
in XML documents and then (if needed) in VRML
files trough i.e. XSL transformers. This
component will provide a navigable VRML data and,
as added value, a textual XML document the
extensible Mark-up Language is the key for the
standardization and the media of the data. The
XML format allows to represent these data in a
lot of different structures to underline some
aspects above others.
ASI3D Data analyzer
XML data
ASI3D XSL transformer
VRML data
ASI3D Virtual scenario
14
MASS Context and ASI-3D

• The MASS project aims to to build many end-user
  services based on earth observation. The services
  will cooperate through an XML data comunication
  infrastructure. In this way the system will
  provide Earth Observation (EO) Services possibly
  integrated with meteorological and GIS data
  through the exploitation of XML related protocols
  and standards.
• The use of the ASI-3D approach should give the
  possibility to integrate MASS approach to
  applicative infractures capable to process large
  data-sets in parallel and with scalable
  performances, to provide EO data in the web based
  infrastructure.