Technical Tips

This technical tip briefly explains how and why Spatial Vision used Virtual Reality Modelling Language (VRML) to generate web-viewable interactive 3D models of Victorian state-wide Groundwater Management Units (GMUs).

Download VRML-Plugin-Blaxxun_Contact_10487.zip (For Internet Explorer v4+) (5.5mb)

Introduction

Over recent years there has been a rapid increase in interest in the display of three dimensional (3D) geospatial data over the Internet; most notably with the release of Google Earth in 2005. For many, 3D visualization can be a powerful means to communicate complex information or to make a broad number of people familiar with a situation quickly. Numerous tools for 3D Web visualisation are now available to be employed.

In a recent project, Spatial Vision was requested to provide a visualisation tool to communicate the relationships between Groundwater Management Units (GMU) to scientists, stakeholders and the wider community. GMUs are complex and almost intangible subterranean units used in the management of groundwater that are physically impossible to see. 3D visualisation offered a means to better understand the physical size of the GMU, their relationship to one another and to other topographic features and localities. Another significant benefit of visualisation was to provide a means to verify the integrity of the available spatial data about each GMU.

Figure 1: 3D interactive models of Groundwater Management Units

VRML-HTML-JavaScript Solution

Spatial Vision used a combination of Virtual Reality Modelling Language (VRML), HTML and JavaScript to produce a web-based application to visualise and interact with the 3D model. This technology combination enabled Spatial Vision to deliver the 3D model with tools and supporting information in an easy to use interface.

As well as displaying the groundwater management information, the 3D model includes overlays of satellite imagery, major watercourses, major road networks and town locations and their areal coverages. The 3D data was initially generated in the GIS modelling application, ESRI's ArcScene, which was then exported to VRML format (.wrl), where it was edited and optimised for visualisation within the model. Text, links to database records and other HTML documents were also added to complete the model before it was tested and refined.

Virtual Reality Modeling Language (VRML)

The focal point of this solution was the 3D delivery system based on VRML (Virtual Reality Modelling Language). VRML is an open source language, developed by the Web3D Consortium (http://www.web3d.org). It is a common export format from many 3D graphics software applications and provides a versatile tool that can be easily customised for different projects. VRML is currently the de facto open source Web3D language but in the future, will be superseded by X3D (eXtensible 3D).

Paradigm for the 3D Model

As described above, the key aim of the visualisation was to improve the gereral understanding of the 3D relationship between GMUs. In order to facilitate this a “layer cake” approach was adopted which comprises a bottom layer showing the surface (Landsat Image), a “floating” middle layer for the GMUs (and associated “height guides”) and a top layer of the ground floating above the GMUs. The three layers where spatially linked by using major towns as reference points. This layering approach helps users to discern the spatial relationships between GMUs.

This paradigm was initially implemented in ArcScene. The layering effects were achieved by manipulation of the key available 3D components:

(1) Base Height - the reference height used to locate a layer;
(2) Offset - a variable that enables an entire data layer to be moved up or down (in our case floated above) by a set distance;
(3) Extrusion - the variable that allows a 2D shape to be extended into the Z dimension by some set value (which in this case was the thickness of the GMU), and
(4) Exaggeration - a variable that allows the vertical to be exaggerated with respect to the horizontal (in this case by a factor of 10 was adopted as the default view).

The 3D Scene is then exported to VRML for final editing and incorporation into the virtual reality framework set-up to accommodate the multiple views for this project.

How to Use the GMU Visualisation Tool

The GMU visualisation tool can be found and used directly over the web, see http://www.spatialvision.com.au/html/GMU_VRML/index.html. The VRML model can be viewed using the Web browser Internet Explorer 4 or higher and requires the free Blaxxun Contact Web browser plug-in (www.blaxxun.com/home/index.php).

Upon start up the user is presented with an HTML and JavaScript home page (see Figure 2). This start up screen provides information about this visualisation tool and requirements to run it. If another VRML plug-in is running on the computer, it must be removed and Blaxxun Contact 5.3 must then be loaded. Once the appropriate set-up has been achieved, the user can then select one of the six groundwater zones to view in 3D. This is done by simply clicking on the desired zone on the map of Victoria.

The view of GMUs shown is current as at October 2007. However, GMUs for the Victoria are subject to ongoing review and the views presented are for demonstration purposes only.

Figure 2: Home page for the GMU visualisation tool

The user is then presented with one of the six main visualisation pages (one for each of the six regions of the state as identified in the state map of Victoria) as shown in Figure 3 below.

Figure 3: North East zone - one of the six groundwater zones

Initially, navigation around the 3D model can be awkward but the user should become satisfactorily proficient at movement control with 15-30 minutes practice. To assist the user, model-specific navigation controls and set viewpoints are available on the right side frame. Navigation options are provided in the form of the "Rotate", "Zoom", "Fly" and "Pan" buttons, which work as follows:

Rotate - Press the left mouse button to select a rotation point within the scene. Then drag the mouse left right up or down to rotate the object.
Zoom - Click with the mouse on object to move toward that point.
Fly - Press the left mouse button to start flying. Drag the mouse toward the top or bottom to rise or sink, drag toward the left or right to turn.
Pan - Drag the mouse to move up down left or right within a plane which is perpendicular to your viewing direction.

Figure 4 presents one of the many possible results of manipulating the 3D model. In the view presented, the user has turned off the surface “satellite image” and “height guides” layers, turned on the “GMU names” layer, then by using the navigation controls zoomed into a region of interest and then queried the Barnawartha GMU by simply clicking on it. Clicking on the GMU presents a pop-up window of information about the GMU.

Figure 4: One of the many possible results of manipulating the 3D models

Summary

The sharing of 3D geospatial data via the Web is a rapidly growing area as demonstrated by the success of Google Earth. The VRML approach used in this project proved to be a very effective tool for interpreting and disseminating complex 3D information. For future reference, X3D (eXtensible 3D) has been earmarked by the Web3D Consortium as the successor of VRML. Support for X3D will solve some of the restricting issues associated with this project, such as being tied to only one Web browser and one VRML plug-in option. X3D also includes support for more geospatial data formats and new 3D feature types (such as 3D volumes and clipping planes). More information on X3D is available from the Web3D Consortium Web site: http://www.web3d.org (Web3D Consortium, 2007).

back to top

Copyright © Spatial Vision, Wednesday, 31-October-2007