`Bradium Technologies LLC - Patent Owner
`Microsoft Corporation - Petitioner
`IPR2016-00448
`
`1
`
`
`
`ups ARCHIVE
`1998'o9
`LEAVER, R.
`
`__
`I
`.'
`
`'
`
`Z"
`
`_-
`-
`
`I
`
`I"
`
`.
`" I
`
`.
`
`I
`
`I
`
`.
`
`“‘
`
`_~
`H"
`
`.-
`5 .-1'-'.-:':*..'-"4*.;*.:f.r
`'
`*
`‘
`
`.5. __r-;
`
`2
`
`
`
`‘I1 :1 ELEV KNOX :-atsh.£‘~m’
`-..-M Pi‘...-‘."{:D,Af‘HJ.-\T"
`
`"-1
`
`-
`
`'Ir~.'Tr?nE\
`
`r‘.'=.
`
`;
`
`3
`
`
`
`4
`
`
`
`5
`
`
`
`NAVAL POSTGRADUATE SCHOOL
`
`Monterey, California
`
`
`
`THESIS
`
`VRML TERRAIN MODELING FOR THE MONTEREY
`
`BAY NATEONAL MARINE SANCTUARY (MBNMS)
`
`by
`
`R. Greg Leaver
`
`September 1998
`
`‘
`J
`
`Thesis Advisor:
`Associate-Adviser:
`
`Don Brutzunan
`Rex Buddenberg
`
`
`
`Approved for public release; distribution is unlimited.
`
`6
`
`
`
`7
`
`
`
`
`
`
`
`
`ORT OCUMENTATIONAPAG is
`M Y mm
`
`Public repu-rtrn:_: burden for this collcetiug of mtonnatron I5 esttrnated to average l hour per rr~.pmist-: Including the ttmdtul teuciung ll'I>'lIUC'l1(Jfl.
`searching existing data sources. gathering and maintaining the data needed. and completing and reviewing the collection of information. Send
`
`l
`comments regarding this burden estimate or any other aspect ofthis collection of inl'nr1'nation, including suggestions for reducing this burden. to
`Washington headquarters Services. Directorate for Information Operations and Reports. I2 I5 Jefferson Davis Highway. Suite 120-4. Arlington. VA,
`Illa].--ljol. and to the Lillie: cl .'.l..1.t'.ttgcu..:nt and L-u..1g:.-L l-’.1p.:m.;rl\' Ix;-Juctiolt Project tuT.J'+ l.1.'.t$: Vvd.1l'llugluIlLK_' loft.
`
`Faro: _-lppromd
`
`
`
` 1. AGENCY USE ONLY 2. REPORT DATE
`
`3. REPORT TYPE ,\.\u'D DATES COVERED
`
`
`
`
`
`
`September 1998
`
`Master’s Thesis
`
`4. TITLE AND St‘-BTITLE : VRML Terrain Modeling for the Monterey Bay National
`Marine Sanctuary (MBNMS)
`6. AUTIIOR
`
`5_ FUNDING _\.UMBERs
`
`R. Greg Leaver
`
`7. PERFORMING ORGANIZATION NAMEIS) AND Aoottssstesi
`Naval Postgraduate School
`Monterey. CA 939436000
`
`9. SPONSORING I !\IONlTORl.VG AGE."\'CY .\EAl\1E[5)AI\"D ADDRESSHZS)
`NIA
`
`11. SUl’PI.E.\lENTAR‘:’ NOTES
`
`%R*;L£,:fE‘:'¥I'3E REPORT
`NUMBER
`
`I0. SPONSORING!
`MOl\'lTORl'l\'G
`AGENCY REPORT
`NUMBER
`
`The views expressed in this thesis are those ofthe author and do not reflect the official policy or position of the
`Departrnent of Defense or the US. Government.
`12a. DISTRIBUTION!AVA[LABILIT\'STATEMEl"4T
`1 Approved for public release; distribution is uniirnited.
`_'
`13. ABSTRACT (ma.!'imTtm 200 words)
`‘
`This thesis develops an online model of the topographic terrain of Monterey Bay National Marine Sanctuary
`L (MBNMS) seafloor. Written in the Virtual Reality Modeling Language (VRML), the model is an interactive 3D
`application composed of hundreds oftopographic tiles linked together to form a mosaic of the bay. Low-resolution
`tiles are traded for higher resolution tiles as the viewer gets closer to the terrain.
`
`12b. DISTRIBITTION CODE
`
`test usage of
`proposed metadata conventions linking VRML and the extensible Markup Language (XML), demonstrated use of the
`GeoVRl\/TL Working Groups proposed QuadLOD node, and a preliminary 3D navigation icon for terrain interrogation
`and wayfinding. Terrain data was produced from registered, smoothed and subsampled bathvmetric sonarscan
`results. Because the model is geo-re ferenced with the Universal Transverse Mercator (UTM) coordinate system, a
`user can easily add scientific content or data to a selected location ofthe MBNMS in a manner analogous to adding
`I 2D content to an HTML page. Thus, the user can place 3D content anywhere in the MBNMS in geographic context
`
`
`merely by specifying the geographic coordinates and depth of the content in standard V l~’..\1L 53 r.t:ot.
`
`
`Future work includes improvement of metadata interoperability, navigation icon user testing, and
`autogeneration of image-based texture tiles for scientific visualization.
`14. SUBJECT TERMS
`.
`.
`.
`.
`.
`.
`,
`.
`World Wide Web, Virtual Reality Modeling Language WRML), Large—Scale Virtual Environments
`(LSVES). Monterey Bay, 3D Graphtcs Modeling
`
`
`
`
`
`15. \'Ul\lBER
`OF :1AGES
`126
`
`
`
`
`
`
`CODE
`
`I.
`
`-
`
`
`\
`1
`’
`
`
`
`
`
`‘
`
`_L
`
`-
`
`\
`
`IE1
`
`\
`
`u
`
`5- E T H‘-
`
`I
`
`
`
`‘
`
`iii. Ssib‘Cl‘Ht::lno\:0 ‘RE oar
`LA suit AT!
`.
`I-
`P
`Unclassified
`
`Nsx ‘.-5.1{|-[]].23[|-55m3
`
`‘I38? ilhgfsbiiigi mm“ A [Or] i:'9LaSshs(iiiiigiii|os or
`;*[i""T‘”'°N
`
`Unclassified
`.
`.
`OF ABSTRACT
`A35'm—"CT
`
`
` Unclassified UL
`Standard Form 298 (Rev. 2-89)
`Prescribed by ANSI Std. 239-18
`
`8
`
`
`
`ii
`
`9
`
`
`
`Approved for public release; distribution is unlimited
`
`VRML TERRAIN MODELING FOR THE MONTEREY BAY T\".~\TIONAL
`
`MARINE SANCTUARY (MBNMS)
`
`R. Greg Leaver
`Lieutenant. United States Navy
`B.S., Oklahoma State University, 1987
`
`Submitted in partial fulfillment oflhe
`requirements for {he degree of
`
`MASTER OF SCIENCE IN INFORNIATION TECHNOLOGY MANAGEMENT
`
`from the _
`
`NAVAL POSTGRADUATE SCHOOL
`
`September 1998
`
`10
`
`10
`
`
`
`11
`
`11
`
`
`
`ABSTRACT
`
`This thesis develops an online model of the topographic terrain of Monterey Bay
`
`National Marine Sanctuary (MBNMS) seafloor. Written in the Virtual Reality Modeling
`
`Language (VRML), the model is an interactive 31) application composed of hundreds of
`
`topographic tiles linked together to form a mosaic of the bay. Low-resolution tiles are
`
`traded for higher resolution tiles as the viewer gets closer to the terrain.
`
`Important contributions include a naming convention for autogeneration of
`
`interlinked files, test usage of proposed metadata conventions linking VRML and the
`
`eXtensible Markup Language (XML), demonstrated use of the GeoVRML Working
`
`Groups proposed QuadLOD node, and a preliminary 3D navigation icon for terrain
`
`interrogation and wayfinding. Terrain data was produced from registered, smoothed and
`
`subsampled bathymetric sonarscan results. Because the model is geo-referenced with the
`
`Universal Transverse Mercator (UTM) coordinate system, a user can easily add scientific
`
`content or data to a selected location of the MBNMS in a manner analogous to adding 2D
`
`content to an HTML page. Thus, the user can place 3D content anywhere in the
`
`MBNMS in geographic context merely by specifying the geographic coordinates and
`
`depth of the content in standard VRML syntax.
`
`Future work includes improvement of metadata interoperability, navigation icon
`
`user testing, and autogeneration of irnage—based texture tiles for scientific visualization.
`
`12
`
`12
`
`
`
`vi
`
`13
`
`13
`
`
`
`TABLE OF CONTENTS
`
`I.
`
`INTRODUCTION ............................................................................................ ..
`
`1
`
`A. BACKGROUND ......................................................................................... ..
`
`B. MOTIVATION ............................................................................................ ..
`
`D. THESIS
`
`1
`
`1
`
`la)
`
`11.
`
`BACKGROUND AND RELATED WORK ................................................. .. 3
`
`]. Monterey Bay National Marine Sanctuary ....................................... ..
`
`2. Monterey Bay Modeling
`
`DJ
`
`3
`
`5
`
`3. Coordinate SystemsUsed 5
`
`4. What is
`
`1. GeoVRML Working
`
`2 Seamless Solution‘s Terrain
`
`3. SIGGRAPH CARTO Project ............................................................ ..
`
`4. VRML Terrain
`
`6
`
`7
`
`8
`
`8
`
`8
`
`5. Synthetic Environment Data Representation 3: Interchange
`
`Specification (SEDRIS) ..................................................................... .. 9
`
`6. Other Existing VRML Terrain Models ............................................. .. 10
`
`III.
`
`PROBLEM STATEMENT ............................................................................ .. 13
`
`vii
`
`14
`
`
`
`A.
`
`INTRODLICTION ....................................................................................... ..
`
`13
`
`B. RESEARCH FOCUS .................................................................................. ..
`
`I3
`
`C. DESIGN CONSIDERATIONS .................................................................... ..
`
`I. TransitionsConsidered
`
`a. "SwapTi1e" Transition ........................................................... ..
`
`b. "QuadTiIe"
`
`c. "QuadSwapTile"
`
`14
`
`I5
`
`16
`
`16
`
`16
`
`2. Transition Chosen ............................................................................. ..
`
`17
`
`18
`
`IV.
`
`BATHYMETRIC TERRAIN DATA.............................................................. ..
`
`I9
`
`A. INTRODUCTION ....................................................................................... ..
`
`19
`
`B. DATA
`
`1. Data Source and Gridding
`
`2. Partioning the Datasets ....................................................................... ..
`
`3. How Resolutions Were Determined
`
`C. FILE NAMING
`
`20
`
`22
`
`22
`
`V.
`
`JAVA PROGRAMS FOR DATAFILE CONVERSION TO VRML .......... ..
`
`25
`
`B. CREATEVRMLTILE PROGRAM: GENERATING INDIVIDUAL
`
`I. Read Data FileName
`
`26
`
`2. Read Metadata
`
`26
`
`viii
`
`15
`
`15
`
`
`
`3. Read Elevation Data ......................................................................... .. 28
`
`4. Geographically Position Tile ............................................................ ..
`
`5. Write VRML Syntax ......................................................................... ..
`
`28
`
`28
`
`C. CREATEVRMLTREE PROGRAM: GENERATING LINKING
`
`VRML TERRAIN TREES .......................................................................... ..
`
`28
`
`1. Read Children File Names ................................................................. ..
`
`2. Construct Parent and Children Relationship ...................................... ..
`
`3. Write VRML Syntax .......................................................................... ..
`
`29
`
`29
`
`29
`
`D. SUMMARY ................................................................................................. ..
`
`29
`
`VI.
`
`VRML SCENE DETAILS .............................................................................. ..31
`
`B TERRAIN TILES .......................................................................................... ..
`
`31
`
`3]
`
`1. Metadata .............................................................................
`
`............. ..
`
`31
`
`2. Positioning ......................................................................................... ..
`
`3. Navigation Icons ................................................................................ ..
`
`34
`
`4. ElevationGrids
`
`35
`
`3. Textures .............................................................................................. ..
`
`36
`
`2. Viewpoints
`
`36
`
`37
`
`D. SUMMARY ................................................................................................. ..
`
`38
`
`VII.
`
`EXPERIMENTAL RESULTS ........................................................................ ..
`
`39
`
`B. MONTEREY BAY TERRAIN MODEL DATABASE
`
`39
`
`ix
`
`16
`
`16
`
`
`
`C. PIERFORMANCE RESULTS AND USABILITY TESTING ..................... .. 40
`
`1. Performance Aids ............................................................................... .. 41
`
`:1. Vertical Exaggeration .................................................................... .. 41
`
`1). Reducing File Size by Rounding .................................................. .. 42
`
`2. Perfomiance Results .......................................................................... .. 42
`
`D. USER ACCESS AND NAVIGATION ........................................................ .. 43
`
`E. EXAMPLE INTEGRATION OF CONTENT ............................................. .. 45
`
`l. Georeferencing ................................................................................... .. 45
`
`2. Adding Content .................................................................................. .. 46
`
`F. SUMMARY .................................................................................................. .. 47
`
`VIII. CONCLUSIONS AND RECOMMENDATIONS ......................................... ..-49
`
`A. RESEARCH CONCLUSIONS .................................................................... .. 49
`
`1. Generating VRML Syntax ................................................................. .. 49
`
`2. Viewpoints
`
`49
`
`3. Rendering ........................................................................................... .. 50
`
`4. Georeferencing and Content .............................................................. .. 51
`
`B. RECOMMENDATIONS FOR FUTURE MBNMS TERRAIN
`
`1. Normals to Eliminate Tile Seams ...................................................... .. 51
`
`2. Modified QuadLOD Node ................................................................. .. 52
`
`3. Navigation Icons to Control Other Transitions.................................... 53
`
`4. Return ofthe Navigation Icons
`
`54
`
`5. Other MBNMS Model Future Work ................................................. .. 54
`
`C. OTHER LSVE FUTURE WORK ................................................................ .. 54
`
`X
`
`17
`
`17
`
`
`
`APPENDIX A: SCRIPTS USEDTO GRID DATA SETS ....................................... ..57
`
`APPENDIX B: SCRIPT USED TO PARTITION DATA SETS ........................... ..6I
`
`APPENDIX C: CREATEVRJVILTILE JAVA PROGRAM ................................... ..63
`
`APPENDIX D: CREATEVRMLTREE JAVA PROGRAM .................................. ..77
`
`APPENDIX E: EXAMPLE VRNIL TERRAIN TILE FILE STRUCTURE ......... ..8I
`
`APPENDIX F: EXAMPLE VRML TERRAIN TILE SCENE GRAPH ............... "89
`
`APPENDIX C: EXAMPLE XNIL FILE ................................................................... ..93
`
`APPENDIX I-I: EXAMPLE VRML TERRAIN TREE FILE STRUCTURE ....... ..95
`
`APPENDIX I: EXAMPLE VRML TERRAIN TREE SCENE GRAPH .............. ..97
`
`LIST OF REFERENCES ............................................................................................ ..99
`
`INITIAL DISTRIBUTION LIST .............................................................................. ..l0l
`
`xi
`
`18
`
`18
`
`
`
`xii
`
`19
`
`19
`
`
`
`LIST OF FIGURES
`
`2.1 Monterey Bay National Marine Sanctuary. .......................................................... .. 4
`
`2.2 Focus of Geo\/RML Working Group ................................................................... .. 7
`
`DJ
`
`D)
`
`DJ
`
`.1 Tile Transitions ..................................................................................................... .. 1 3
`
`.2 Tile Transition Types ............................................................................................ ..17
`
`.3 Implementation of QuadSwapTile Transition
`
`4.1 File Naming
`
`5.] Typical Metadata Excen From Gridded Text Data File ....................................... ..27
`
`6.1 Example XML Keys and Key Values Used in Metadata Node..............................33
`
`6.2 ANavigation
`
`6.3 A Single Elevation Grid (File N3 5361IJ.W1232629.070.05l.1000.seabeam.m'l) .35
`
`6.4 QuadLOD Excerpt (File N3536l0.\Vl232629.070.05l.1000.tree.wrl)
`
`7.1 Directory Structure of MBNMS Terrain Model Database.
`
`7.2 Applying Vertical Exaggeration
`
`7.3 3:] Scaling Example (File N3536] 1.W1223609.070.0Sl.l00O.seabeam.wrl) ..... ..4l
`
`7.4 1:1 Scaling Example (File N3536l 1.Wl223609.070.051.l000.Seabeam.wrl) ..... ..42
`
`7.5 Entry Viewpoint OFMBNMS TerrainModel.
`
`7.6 Georeferencing VRML and UTM Coordinate Systems
`
`xiii
`
`20
`
`
`
`xiv
`
`21
`
`21
`
`
`
`LIST OF TABLES
`
`4.1 File Partition Characteristics by Resolution for Entire MBMNMS F0otprint...........21
`
`6.1 Quadlod Node Proximity Sensor Values. ................................................................ ..37
`
`7.1 Terrain Tile Dataset Characteristiics ....................................................................... ..40
`
`7.2 Terrain Tree Dataset Characteristics ........................................................................ ..-40
`
`7.3 Rendering Time for Resolutions .............................................................................. ..43
`
`7.4 Tile Switching Values (depth) .................
`
`............................................................. ..45
`
`xv
`
`22
`
`22
`
`
`
`xvi
`
`23
`
`23
`
`
`
`ACKNO\VLEDGEMENTS
`
`To my wife Sandra. my daughter Lauren, and my son Chase. thank you all for
`
`supporting me in this endeavor.
`
`I love and treasure each of you more than mere words
`
`can say. To Don Brutzman, I offer my thanks for your contagious inspiration,
`
`enthusiasm, and guidance. To Ray McClain, I am indebted to you for your assistance in
`
`this project.
`
`I couldn"'I have done it without your help.
`
`xvii
`
`24
`
`24
`
`
`
`25
`
`25
`
`
`
`I.
`
`INTRODUCTION
`
`A.
`
`BACKGROUND
`
`This thesis investigates how the Virtual Reality Modeling Language (VRML) can
`
`be used to model the seafloor topography of the Monterey Bay National Marine
`
`Sanctuary (MBNMS). By creating a topographic model of the MBNMS using VRML, a
`
`three-dimensional representation of the SaI'lC‘t'l.l3.I'}r’ can be accessed over the World-Wide-
`
`Web (Web) by anyone using a VRML-enabled web browser or standalone VRML
`
`viewer. A VRML-enabled browser means a browser configured with a VRML plug in
`
`such as Cosmo Player for PCS (Silicon Graphics, 98). Rapid recent progress in this field
`
`means that many new opportunities are available.
`
`B.
`
`MOTIVATION
`
`Numerous scientists and researchers are collecting data and building
`
`environmental models about Monterey Bay. Regional research partnerships using a
`
`Large Scale Virtual Environment (LSVE) for Monterey Bay will make it easy for
`
`scientific content about Monterey Bay to be placed and accessed online. Building a
`
`Monterey Bay terrain model is a dramatic way to encourage scientists to their work in
`
`three-dimensional (3 D) space and on the Web. New insights and new research
`
`collaborations are likely. A new paradigm for publication of scientific data and analytic
`
`results is possible.
`
`C.
`
`OBJECTIVES
`
`The goal is to make the addition ofa user-selected portion of MBNMS terrain in a
`
`3D VRML scene as easy as adding a background image to a 2D HTML page. Thus, it is
`
`26
`
`26
`
`
`
`hoped that this effort will make it easy for scientific content about Monterey Bay to be
`
`placed and accessed online, in a 3D geographic context.
`
`Although many scientists are conducting research in Monterey Bay, bathymetric
`
`terrain scenery is not easily available.
`
`— Constructing a LS VE for Monterey Bay may dramatically enhance ongoing
`
`regional research collaborations. An additional objective is for the model to support
`
`variable resolutions of gridded data. "Variable resolutions" essentially means that as a
`
`viewer gets closer to terrain, the resolution of the terrain increases to provide superior
`
`granularity.
`
`D.
`
`THESIS ORGANIZATION
`
`The remaining chapters of this thesis are organized as follows. Chapter 11
`
`provides the background for the effort, introduces a few cartographic concepts, and
`
`touches on some related work being done. Chapter III presents the problem statement
`
`and covers design considerations for a feasible solution. Chapter IV provides a look at
`
`bathymetric data sources and describes the gridding process used to create simple gridded
`
`text data files. Chapter V shows how these elevation grid terrain text files can be
`
`processed by a Java program to produced VRML world files. Chapter VI discusses the
`
`specific VRML constructs produced by the Java programs that implement the MBNMS
`
`terrain model. including 3D navigation/information icons. Chapter VII considers
`
`experimental results, examines user access, and shows how users can integrate their 3D
`
`content and MBNMS terrain. Chapter VIII presents thesis conclusions and provides
`
`recommendations for future work.
`
`27
`
`27
`
`
`
`II.
`
`BACKGROUND AND RELATED VVORK
`
`A.
`
`INTRODUCTION
`
`This chapter examines pertinent background work that motivated the construction
`
`ofa terrain model for the MBNMS and introduces the basic concepts of VRML.
`
`It also
`
`discusses other work being done to produce 3D topographic models, and provides a quick
`
`look at a tool evaluated by the author that creates 3D topographic scenery in VRML from
`
`elevation data sets.
`
`B.
`
`BACKGROUND
`
`1.
`
`Monterey Bay National Marine Sanctuary
`
`Beginning 1] km north of San Francisco's Golden Gate Bridge, the MBNMS]
`
`extends 260 km south along the California coast to Cambria Rock in San Luis Obispo
`
`County. The Monterey Bay National Marine Sanctuary contains the nations greatest
`
`diversity of marine life and habitat. East to west, the sanctuary stretches 152 km and
`
`holds one of the world's largest ocean canyons: the 10,663 ft. deep Monterey Canyon.
`
`Thus, this area provides unparalleled opportunities for marine scientists based at nearby
`
`research institutions such as Moss Landing Marine Laboratories (MLML), Monterey Bay
`
`Aquarium Research Institute (MBARI), and NPS. Figure 2.1 illustrates the MBNMS
`
`region. The sanctuary was established to enhance resource protection and preserve the
`
`natural beauty within its boundaries.
`
`1 More information on the MBNMS is available at hug.‘//bonira.mbnm5.n0s.noaa.govf
`
`28
`
`28
`
`
`
`29
`
`29
`
`
`
`275 km
`
`220 km
`
`l65km
`
`ll0km
`
`55 km
`
`OI-cm
`
`170 km
`
`135 km
`
`90 km
`
`45 km
`
`0 km
`
`Figure 2.1. Monterey Bay National Marine Sanctuary (MBNMS Web Site, 98)
`
`30
`
`30
`
`
`
`31
`
`31
`
`
`
`2.
`
`Monterey Bay Modeling Group
`
`Interest in developing computerized processes and models to assist with studying
`
`and managing the sanctuary led to the formation of the Monterey Bay Modeling Group.
`
`In 1993. the MBNMS Research Advisory Committee. under the sponsorship of the
`
`National Oceanic and Atmospheric Administration (N OAA), prepared a research plan
`
`which outlined the research priorities and management goals for the sanctuary. This plan
`
`outlined the objectives of the Monterey Bay Modeling Group, an ad hoc group of
`
`individuals interested in computer modeling and affiliated with various MBNM S research
`
`organizations, including NPS. Listed among the objectives was the goal for the
`
`development of a computerized model of the sanctuary (NOAA, 93). The model, it was
`
`hoped, would ultimately function as an oceanographic scientific database archival and
`
`retrieval system, which could be overlain on a 3D physiographic representation of the
`
`MBNMS. The model would be networked for use by scientists, engineers, planners,
`
`managers, and the general public. Unfortunately this group was only active for two
`
`years. Recent discussions indicate that technology has advanced sufficiently to enable
`
`further scientific collaborations.
`
`3.
`
`Coordinate Systems Used
`
`In terms of latitude and longitude, the MBNMS occupies a square region between
`
`35°3 0' North and 38° North latitude, and 123315‘ East and 121“ East longitude. Since
`
`latitude and longitude are commonly used measures. they are included in the model's
`
`metadata and file-naming convention. Universal Transverse Mercator: (UTM)
`
`coordinates, which specify a location as a distance north (Northing) and east (Easting)
`
`2 A good explanation of the UTM coordinate system is available at: hrtg.'/Cdgeograghv.{gum/msub}4.h!m
`
`32
`
`32
`
`
`
`from :1 zones meridian measured in meters. are also frequently used in cartography.
`
`In
`
`terms ofthe UTM system. the MBN MS lies between Northing coordinates of
`
`3.940.000m to ~'l.200.000m, and Easting coordinates of 460.{)D0m to 6l2_.00{]m. Because
`
`L|'l'M coordinates are in meters rather than degrees. UTM measurements can easily be
`
`converted to VRML coordinates that are default units in meters. This capability allows
`
`gridded elevation data and scientific content to be positioned relative to their real world
`
`location in the MBNMS model.
`
`4.
`
`What is VRML?
`
`VRML - the Virtual Reality Modeling Language — is a 3D graphics scene
`
`description language that enables a scene builder to create dynamic worlds and sensor
`
`rich virtual environments on the Internet. VRML enables users to animate objects in
`
`worlds, making them move; it also enables users to play sounds within worlds, interact
`
`with worlds and, control and enhance worlds with scripts, or small programs (Ames, et.
`
`al., 97). VRML provides a standardized, portable, and platform-independent way to
`
`render dynamic, interactive, 3D scenes across the Internet (Brutzman, 97).
`
`A VRML file generally ends with extension ".wrl". This file is a textual
`
`description ofa 3D world. A VRML file contains nodes that describe shapes and their
`
`properties in the virtual world. These nodes make up the building blocks - VRML
`
`constructs - which create the 3D scenery in a virtual world. For cartographic models
`
`such as this project, one of the principal VRML constructs is the EIevan'cmGrz'd node,
`
`which can be used to create a 3D representation of the terrain. The terrain itself is
`
`described by a data set containing bathyrnetric depth values. Each sampled depth value is
`
`33
`
`33
`
`
`
`associated with a pair of gridded 2D coordinates. An excellent overview ofhow VRMI.
`
`can be applied to cartography can be found in Fairborn and Parsley (97).
`
`C.
`
`RELATED WORK
`
`1.
`
`GeoVRML Working Group
`
`To provide a forum for discussions of the representation and exchange ofproperly
`
`geo—referenced data in VRML advance, the GeoV RML Working Group was established.
`
`One of the forum's goals is to establish VRML as a standard for the representation and
`
`exchange of 3D geographic and cartographic data. The GeoVRML mailing list is
`
`maintained as part of the GeoVRML Working Group of the VRML Consortium by SR1
`
`lntemational. Figure 2.2 shows the primary issues of interest and concern of the working
`
`group.
`
`Coordinate systems - measurement systems including the Geodetic and
`Geocentric systems used to specify locations on the surface of the Earth.
`
`Time referencing - important for content that is timestamped with respect to a
`an absolute reference.
`
`Terrain representation - imagery usually represented in an array of numbers
`that represent topography in digital form.
`
`and interoperability.
`
`Levels of detail - the hierarchy of resoltions necessary to achieve acceptable
`rendering and performance for a LSVE
`
`Resolution and accuracy - factors limited by georeferencing VRML worlds to
`a coordinate system and by data storage issues.
`
`Data interchange - standardized data format and type to enable data exchangge
`
`Figure 2.2. Focus of Geo\/RML Working Group (lverson, 98)
`
`2.
`
`Seamless Solution's Terrain Navigator
`
`The Terrain Navigator is implemented entirely in Virtual Reality Modeling
`
`Language (VRML) for use on a low-cost Personal Computer (PC) to enable a content
`
`developer to integrate highly realistic terrain content with Web pages. This real-time
`
`‘-\-.'l
`
`34
`
`34
`
`
`
`interactive visualization software is especially useful for collaborative review ofdatabase
`
`development throughout the design cycle or for entertainment purposes. (Seamless
`
`Solutions. 98).
`
`3.
`
`SIGGRAPH CARTO Project
`
`The "Carto Project" began in 1996 as a cross-organizational collaboration
`
`between the activities of the Association for Computing Machinery's Special Interest
`
`Group on Graphics (ACM SIGGRAPH) and the International Cartographic Association's
`
`(ICA) Commission on Visualization. The Cane Project explores how viewpoints and
`
`techniques from the computer graphics community can be effectively applied to
`
`cartographic and spatial data sets. This includes exploring how viewpoints and methods
`
`from cartography can enhance developments in computer graphics; especially those
`
`associated with the representation of geographic phenomena. These efforts will continue
`
`into 1999, in conjunction with the time frame of the ICA's Commission on Visualization
`
`(Rhyne, 98).
`
`4.
`
`VRML Terrain Generators
`
`Several commercial products exist that can automatically generate VRML terrain.
`
`To do so, generally these products import a dataset in a prescribed format such as Digital
`
`Elevation Model3 (DEM) and produce export a VRML file via a filter. Rapid Imaging
`
`Software offers a product called LandForm Gold (RIS, 98) that works like this. A copy of
`
`this software was evaluated by this author, courtesy of Mike Abemathy at RIS.
`
`. LandForm Gold is a powerful 3D real-time terrain viewer for the Windows NTx’9S
`
`platform. This product allows a user to view geographical data in a three-dimensional
`
`3 DEM files contain data of the elevation of the terrain over a specified area, usually at a fixed grid
`
`35
`
`35
`
`
`
`representation and move through the data in a natural and intuitive manner. The program
`
`accepts numerous file types and allows the user to superimpose an image of the area over
`
`the terrain. This effect ofthe image overlay combined with 3D data creates a strikingly
`
`realistic representation of the terrain, as landmarks and topographical features are
`
`dramatically revealed in 3D. As mentioned earlier. La.ndForm Gold also enables the user
`
`to create VRML models based upon the dataset read by the viewer. Other tools, such as
`
`Cybertrelt (98) and Coryphaeus (98) are also available to create VRML terrain models,
`
`but were not evaluated by this author.
`
`5.
`
`Synthetic Environment Data Representation & Interchange
`
`Specification (SEDRIS)
`
`The SEDRJS Geographic Reference Model (GRM) has been proposed by the
`
`GeoVRML Working Group (discussed later) as a standard for VRML coordinate
`
`systems. SEDRJS is a reference model and software package that currently supports 12
`
`different commonly used world coordinate system convention, as well as tools to
`
`automatically convert reference marks between them. Coordinate system standards
`
`supported include Geodetic (GDC or latitudeflongitude), Geocentric (earth centered
`
`Cartesian). Universal Transverse Mercator (UTM), and Lambert Conformal Conic
`
`(LCC). The proposal (GeoVRML, 98) was drafted by SRI International and is
`
`summarized here. It proposes two levels to employ VRML constructs that implement the
`
`SEDRJS standards. Level 1 consists of a means of entering geographical coordinates into
`
`VRML files so that the Cartesian VRML coordinates are generated with respect to a
`
`geographically referenced local coordinate system. Its use depends only on the
`
`interval, such as 1 arc—degree or ?.5 arc-minutes. DEM files are avilable (for a fee) from the US.
`Geological Survey at: htrgsfledcmvw. cr. nsgs. got-mrebglrs
`
`36
`
`36
`
`
`
`availability ofa librar_v for converting from geographical coordinates in the GRM into a
`
`local Cartesian frame. [.evel 2 consists ofan attempt to establish a means for
`
`atltomatically managing the relationships between the local Cartesian frames defined in
`
`Level 1. It is intended as the enabling technology for seamlessly integrating accurately
`
`georeferenced worlds from a wide variety of sources. Since the constructs contained in
`
`the proposal are experimental at this time, they were not employed in the MBNMS
`
`Model. Nevertheless they remain an important area for fiiture work.
`
`6. Other Existing VRML Terrain Models
`
`SRI International has developed a VRML terrain model of the Fort Irwin,
`
`California area. This terrain model has been distributed on CD-ROM and is also
`
`viewable on the Web (SR1. 98). It uses multiple levels of detail to change the terrain‘s
`
`resolution based upon the viewer's distance to the scenery.
`
`In 199'? RIS produced a
`
`model of the San Francisco Bay area (Abernathy, 93). This model was produced to
`
`convey topographical information to participants in the San Francisco Relay. By
`
`integrating Global Positioning Satellite elevation data with satellite and aerial imagery,
`
`the model displayed the terrain the event's course and scenery from a runner's point of
`
`view.
`
`A simple textured model of Monterey regional terrain is also available at
`
`hrrg://ece.uwaterIo0.ca/vrmZ98. It provides background for the VRML 98 Symposium
`
`313 Website
`
`I).
`
`SUMMARY
`
`This chapter explores the related work that preceded or motivated the creation of
`
`a model for the MBNMS.
`
`It places the sanctuary in a geographical context and presents
`
`10
`
`37
`
`37
`
`
`
`the goals of the Monterey Bay Modeling Group. A brief overview of VRML provides
`
`some basic concepts of this scene-description language. Work related to VRML terrain
`
`development is considered and VRML terrain authoring tools are introduced - one of
`
`which is evaluated by the author. Additionally, some existing VRML terrain models are
`
`identified.
`
`ll
`
`38
`
`38
`
`
`
`39
`
`39
`
`
`
`Ill.
`
`PROBLEM STATEMENT
`
`A.
`
`INTRODUCTION
`
`Although excellent commercial software exists for professional development of
`
`topographic models (both VRML and non-VRML) in 3D. these tools can be expensive
`
`and may require data sets to be in a proprietary format. Furthermore- the tools are not
`
`only necessary to generate the topographic models; they are also often required to be
`
`present on a user's console in order to view the models. By representing the model in
`
`VRJVIL, an open solution to the problem of generating topographic data sets and
`
`subsequently viewing them is obtained. Anyone with a web browser and WWW access
`
`can potentially interact with the model. This chapter covers the problem of developing
`
`such an application.
`
`It then discusses the advantages of VRML as the basis for
`
`implementing a solution. Much of the chapter is devoted to an examination of the design
`
`issues considered in the development of a model solution.
`
`B.
`
`RESEARCH FOCUS
`
`In the last few years, advanc