3D LASER SCANNING: ACQUISITION, VISUALIZATION AND ANALYSIS CENTER FOR LITHOSPHERIC STUDIES DEPT' OF - PowerPoint PPT Presentation

Loading...

PPT – 3D LASER SCANNING: ACQUISITION, VISUALIZATION AND ANALYSIS CENTER FOR LITHOSPHERIC STUDIES DEPT' OF PowerPoint presentation | free to download - id: 208dc6-YWVhO



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

3D LASER SCANNING: ACQUISITION, VISUALIZATION AND ANALYSIS CENTER FOR LITHOSPHERIC STUDIES DEPT' OF

Description:

3D LASER SCANNING: ACQUISITION, VISUALIZATION AND ANALYSIS CENTER FOR LITHOSPHERIC STUDIES DEPT' OF – PowerPoint PPT presentation

Number of Views:1266
Avg rating:3.0/5.0
Slides: 213
Provided by: xuemi
Learn more at: http://www.utdallas.edu
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: 3D LASER SCANNING: ACQUISITION, VISUALIZATION AND ANALYSIS CENTER FOR LITHOSPHERIC STUDIES DEPT' OF


1
3D LASER SCANNING ACQUISITION, VISUALIZATION AND
ANALYSISCENTER FOR LITHOSPHERIC STUDIESDEPT. OF
GEOSCIENCESUNIV. OF TEXAS AT DALLAS
  • CARLOS AIKEN 972 883 2450 214 535 6520
  • XUEMING XU 972 883 2405
  • JOHN THURMOND214 718 2110
  • MOHAMED ABDELSALAM 972 883 2724
  • M. ULIA OLARIU
  • ALLIE THURMOND
  • ADMINISTRATIVE ASSISTANT SHARON EDWARDS (CENTER
    FOR LITHOSPHERIC STUDIES) 972 883 2424

2
SATURDAY AGENDA
  • 830AM-- ARRIVE
  • 900AM--AGENDA AND OVERVIEWCA
  • EMPHASIS ON TEAM APPROACH AND CASE HISTORIES
  • 945AM--MODEL BUILDING post-processing, modeling
    and softwares XX
  • 1030AM 3D ANALYSISJT,UO,XX,CA
  • 1100AM BREAK
  • 1115AM CONTINUE
  • 1145AM REVIEW OF FIELD EXERCISE AT PRESTON
    CANYON-CA
  • 1200NOON--LUNCH OFF CAMPUS AT SONNY BRYANS
  • 100PM-- GO TO FIELDPRESTON CANYON
  • 130-330PM-- MAP PRESTON CANYON
  • 4-430PM-- RETURN TO CONVENTION CENTER

3
SUNDAY MORNING AGENDA
  • 830AM-- ARRIVE
  • 900AM-- OUTLINE OF ACTIVITIES
  • 915AM-- PRESTON CANYON-IN 3D STEREO WITH GEOWALL
  • FIELD DATA ACQUISITION
  • DATA PROCESSING
  • MODEL BUILDING
  • ANALYSIS OF 3D PRESTON MODELJT,MA,XX,CA
  • 3D MODELING OF STRUCTURE-XX,JT,MA,CA
  • 1100AM BREAK
  • 1115AM CONTINUE
  • 1200-- NOON LUNCH ON SITE

4
SUNDAY AFTERNOON AGENDA
  • 100PM OUTLINE
  • 115PM--CASE HISTORIESin stereo using GeoWall
  • BIG ROCK QUARRY, ARKANSAS-UO
  • SPAIN-JT,XX
  • MT. RUSHMORE-CA
  • MINERAL WELLS (Dobbs Valley)-CA,XX,JT
  • SALINA TUNNEL-JT
  • OTHER-CA,JT
  • 300PM-- BREAK
  • 315PMGENERAL DISCUSSION-ALL
  • 400PM-- WRAP UP
  • RETURN TO CONVENTION CENTER

5
Introduction
  • Mapping is the foundation of the geological
    sciences.
  • Three basic elements in geologic map
  • Location
  • Lithologic information
  • Spatial or geometrical relationships
  • Ultimate goals
  • Providing the digital data sets representing the
    real earth so that a virtual earth can be
    simulated on the computer and then virtual model
    displayed and analyzed in virtual reality
    system.

6
CYBERMAPPING Integration of GPS, Reflectorless
Laser Rangefinders, and other sensors.
7
Technology
  • GPS global positioning to a centimeter.
  • Lasers relative positioning of decimeter to
    centimeter.
  • GIS mapping functionalities.

8
GPS Techniques
Comparing accuracy, baseline, and time required
for positions. SPS, PPS autonomous position.
OTFon-the-fly kinematic post-processing.
RTKreal time OTF kinematic..
9
Summary of the approximate accuracy of GPS
positioning versus methods. (Modified from
Featherstone, 1995)
10
RTK Base Station
Radio Transmitter
GPS Base Antenna
11
RTK Rover Station
12
Diffuse reflection for reflectorless laser
rangefinders
13
904 nm diffuse fractional reflections of common
material
14
Common Type of Reflector Laser Rangefinder
A. Laser Atlanta Advantage CI B. Riegl Scout. C.
Leica Binocs. D. Laser Atlanta Advantage CI with
PRO XL DGPS. E. LaserTech Criterion. F.
LaserTech Impulse.
15
Characteristics of common types of laser
rangefinders
ALS MDL theodolite with a Laser Atlanta (LA)
or Riegl laser rangefinder.
16
(No Transcript)
17
(No Transcript)
18
CyberMapping Design Flowchart
19
Main Window Interface of CyberMapping Software
20
3D perspective view of survey data
21
3D perspective view of survey data
22
Laser Mapping System
Laser rangefinder
Encoder
Portable computer
23
REDUCTION TO THE ELLIPSOID
D
h
S
H
N

R Earth Radius
6,372,161 m 20,906,000 ft.
S D x R R h
h N H
  • R

S D x R N H
Earth Center
24
REDUCTION TO GRID
  • Sg S (Geodetic Distance) x k (Grid Scale
    Factor)
  • Sg 1010.366 x 0.99991176
  • 1010.277 meters

25
REDUCTION TO ELLIPSOID
  • S D x R / (R h)
  • D 1010.387 meters (Measured Horizontal
    Distance)
  • R 6,372,162 meters (Mean Radius of the Earth)
  • h H N (H 158 m, N - 24 m)
  • 134 meters (Ellipsoidal Height)
  • S 1010.387 6,372,162 / 6,372,162 134
  • S 1010.387 x 0.999978971
  • S 1010.366 meters

26
COMBINED FACTOR
  • CF Ellipsoidal Reduction x Grid Scale Factor
    (k)
  • 0. 0.999978971 x 0.99991176
  • 0.999890733
  • CF x D Sg
  • 0.999890733 x 1010.387 1010.277 meters

27
Preston Canyon, Austin Chalk, Plano
  • Initial proof of concept of digital mapping.
  • Integration of GPS, lasers and portable
    computers.
  • Digital terrain by GPS and lasers.
  • Real time strike and dip determination.
  • Initial test of oblique photography.

28
Looking west just east of Preston Road, Plano,
Texas
29
Map view of the mapping of layer and faults.
Continuously shooting Topcon total station
30
North side of cut, vertical section mapped by GPS
controlled Topcon reflectorless laser mapping.
31
Failed attempt to drape oblique photo onto
terrain by rubbersheeting with ESRI software.
Orthophoto draped onto terrain.
32
Looking east along cut with draped orthoquad and
laser mapped geology..
33
Utah Muddy Creek Mapping Project
34
Objectives
  • First test case in the real geological
    environment with ARCO Oil and Gas.
  • Digitally mapping sedimentary bounding surfaces
    and faults

35
Paleogeographic And Modern Topographic Map
36
Photomosaic of Growth Faulted Ferron Sandstone
Member
37
Field Data Acquisition
38
Local GPS locations
39
3D Perspective View of Surveyed Points
40
Rotated Perspective View of Ferron Growth Faults
41
Rotated Perspective View of Ferron Growth Faults
42
200 Constrained surface in attempt to surface fit
43
Final 200 surface trend surafce
44
Surface 100 Constraint
45
Final 100 Surface
46
Thickness of the interval between 100 to 200
surfaces
47
Two Key Bounding Surfaces and Terrain
48
Conclusions
  • The digital mapping system provides a way to
    rapidly map geology in three-dimensions.
  • The surface fits to layer 100 and 200 have 4
    degree dips.
  • The resultant three-dimensional digital model
    allows post-field work analysis, measurements,
    and visualization in three-dimensions.

49
Digital Mapping at Corbula Gulch outcrop, Utah
  • Locate 2D and 3D GPR surveys, coreholes, measured
    stratigraphic sections by RTK.
  • Map the beds along the cliff faces by laser
    rangefinders.
  • Interpolate the 3D geometry of the sedimentary
    bodies.
  • Build the 3D geological model for visualization,
    analysis and interpretation.

50
FERRON SANDSTONE LOCATION
  • Sequence stratigraphic framework well
    established
  • Reservoir analog for oil fields in Gulf of
    Mexico and North Sea
  • Good exposure of the vertical cliff faces
  • Flat mesas and an arid environment are ideal for
    GPR surveying

Y
Delta shoreline
Corbula Gulch
Coyote Basin
X
51
(No Transcript)
52
3-D GPR RESOLUTION
GPR vertical resolution 0.5 m
Seismic vertical resolution 15 m
53
GPR METHOD
  • High resolution electro-magnetic method
  • Similar to seismic methods
  • Characterizes a medium by its electrical
    permittivity, k and electrical conductivity, s.
  • Decreasing velocity with depth
  • Depth of penetration proportional to the loss
    tangent

System console
Antennas
Optic fiber
tan(d) s/e0wk, where e0 permittivity in
a vacuum w angular frequency
54
CORBULA GULCH BASEMAP
  • reservoir simulator voxel scale 3-D GPR cubes
  • (51m x 28m and 31m x 27)
  • reservoir grid cell scale
  • (100mx100m)
  • inter-well scale
  • (550mx350m)

55
CORBULA GULCH FACIES MAP
56
INTEGRATING OUTCROP AND GPR DATA
  • Good correlation of lithology and permeability.
  • Good correlation of lithology and velocity.
  • GPR reflections are produced at the surface
    between layers with contrast in electrical
    properties.

57
3-D GPR INTERPRETATION
58
3-D GPR INTERPRETATION
  • High amplitude, continuous, oblique GPR
    reflections.
  • Tuning effects at thin layers interfaces
    resolved with GPR attributes.

59
INCLINED SURFACES MAPS
60
Objectives
  • 3D integration of data sets, including GPR
    surfaces, borehole and cliff face data.
  • Building 3D model for visualization, analysis and
    interpretation.

61
Photomosaics of Cliff Faces
E
W
S
N
62
Laser Surveyed Points
63
2D Topographic Map
64
3D Perspective View of Survey Data Layout
65
CD 1 Top Surface
66
CD 1 Bottom Surface
67
Thickness Contour Map of CD 1
68
3D Model of Major Bounding Surfaces
69
3D Model of Major Bounding Surfaces
70
GPR Cubes and Profiles of Depth Sections
71
Rendering Procedures
  • Curve fitting of the data
  • Generating initial surfaces
  • Installing constraints to honor geologic
    interpretations.

72
Initial surface
73
Honor geologic interpretation
74
Incline 0 border at Surface C
75
(No Transcript)
76
Incline 4 with laser data
Laser data
77
Major Bounding Surfaces-looking northeast
Z X3
78
Major Bounding Surfaces-looking northwest
Z X3
79
Location of the solid model
80
Volume of bounding surfaces
Incl 5
Incl 6
Incl 4
Incl 7
Incl 3
Topography
C
Incl 0
Incl 2
Incl 1
Upper unit
Unit 1
81
Volume of bounding surfaces
Incl 4
Incl 5
Incl 6
Incl 3
Incl 7
Incl 2
Incl 1
Incl 0
Surface C
Upper unit
Unit 1
82
Volume of bounding surfaces
Incl 2
83
Mismatch
84
Lessons Learned
  • Laser Mapping provides an efficient way to map
    the surface geometry, but
  • Photorealistic model preferable because
    interpretation change and different interfaces
    interpreted from photos

85
Mapping Subtle regional Structures in Wyoming
86
Problem
  • Distribution of sandstone in foreland basins
    controlled by structure.
  • Structures are poorly defined because of low
    amplitude.
  • Modern structures are also folded.
  • Use digital mapping approach to document subtle
    changes in elevation related to basin warping.

87
Lower Belle Fourche MemberFrontier Formation, WY
88
Type Log
89
Photo of Mapped bentonitesPowder River, WY
90
Laser Mapping Bentonites
91
Clay Spur Data Points in 3D Perspective View
92
Data Points on Terrain Map
Powder River, Wyoming
93
Elevation Map of Clayspur Bentonite
94
3D Perspective View of Mapped Horizons
95
Isopach Map of Clayspur to Top Frewens
Complex mounded topography reflecting subtle
structure, broadly oriented NW-SE.
96
Regional Isopach of Clay Spur to 6 Bentonite
showing thinning to the northeast.(from
Bhattacharya et al., 2000)
Outcrop Map area
97
Isopach Map of Clayspur to Top Frewens
Complex mounded topography reflecting subtle
structure, broadly oriented NW-SE.
98
(No Transcript)
99
Isopach Map of Clayspur to Top Frewens
Complex mounded topography reflecting subtle
structure, broadly oriented NW-SE.
100
(No Transcript)
101
Conclusions
  • The digital mapping technique was applied in the
    Powder River Basin to map subtle structural
    folding with relatively low (less than 100 m)
    amplitudes in relation to their wavelengths
    (kilometers).
  • The results show a present-day axis oriented in
    the NW-SE direction with about a 2-3 degree dip.
    This may result from Laramide structures or
    Laramide folding superimposed on top of old
    structures.

102
Conclusions (continuing)
  • The present-day structures were removed by a
    simple thickness constraint (flattening).
  • The isopach and cross sections show strata
    thinning to the north and northeast.
  • This result documents the subtle structural
    folding interpreted to control the deposition of
    the sandstone in structural low on existing
    ancient topography as suggested by Bhattacharya
    and Willis (in press).

103
Lessons Learned
  • Laser Mapping provides an efficient way to map
    the surface geometry, but
  • Photorealistic model preferable because
    interpretations change and different interfaces
    are needed to be interpreted from photos

104
Objectives
  • Render the geological outcrop photorealistically
    in three-dimensions
  • Close-range
  • obliquely
  • Capture the entire outcrop in three-dimensions
    and taking back to office
  • Additional 3D mapping and measurements made back
    in office.
  • RESULT a 3D Virtual World
  • Standard methods for vertical angles not valid

105
Building Photorealistic Virtual Outcrop
  • Objectives
  • Previous Work
  • Image Registration
  • Test Case

106
Objectives
  • Rendering the geological outcrop
    photorealistically in three-dimensions
  • Capturing the entire outcrop in three-dimensions
    and taking back to office
  • Additional 3D mapping and measurements back in
    office.

107
Traditional Image Registration
  • Apply the low order polynomial function to map
    the image coordinates into world space.
  • Only appropriate with a perpendicular perspective
    and relatively small relief
  • Poor accuracy

108
Traditional Image Registration
The georeferenced digital geological map
Geological map draped onto terrain surface
The DEM
109
Geometric Measurement
  • Rely on stereo image sequences.
  • Require a large amount of correspondence points.
  • Time consuming.
  • Not good for natural surfaces.

110
Stereo Image Sequence Capture System
111
Image Registration
  • Rely on pinhole model.
  • Project the ray of a point into camera space.
  • Project to image plane by projective transform.
  • Linear initialization and non-linear optimization.

112
Camera Geometry (Pinhole Model)
113
Regional Geology
Study Site
114
TEST Photorealistic Outcrop at Dallas Post
Office
  • Accuracy of a few centimeters
  • Photo registration about 0.7 2.7 pixels.
  • Bring outcrop back to office photorealistically
    in three-dimensions
  • Directly taking measurement on photos in
    three-dimensions
  • Virtual field trip

115
Dallas Photorealistic Outcrop
116
Perspective View of Surveyed Points
Using GPS and total station mapped layers
(yellow), faults (red) and Terrain (white)
117
Terrain Surface
118
Initial Test
  • Several natural marks identified on the photo,
    and surveyed using laser
  • Transformation established using non-linear
    iterative method
  • Those points back projected into photo-space

119
Initial Test Result
120
Photo Mosaic
121
Comparison with Direct Laser Tracing
  • Pink and blue points are bedding surface and
    fault traced by laser
  • Green and blue points are the control points for
    photograph for transformation
  • Points of bedding and faults back projected to
    photo space

122
Re-Digitized Key Beds and Faults
123
Photorealistic Outcrop
Digital 3D outcrop data may imported into an
immersive visualization environment.
Dr. Xueming Xu reinterprets the Austin Chalk at
Norsk Hydro
124
New Cybermapping System
  • Laser rangefinder/Scanner
  • Acquire digital terrain surface
  • High-end and low-end cost (determined by required
    accuracy, resolution , and speed)
  • GPS
  • RTK system provides global correlation for the
    laser offsets
  • Digital Camera
  • Determined by interchangeable lens and CCD size
  • Consumer-based (non interchangeable, small CCD
    size, large lens distortion)
  • Professional-based

125
Reflectorless Robotic System
126

Big Rock Quarry
Location of Study Area
USA, Arkansas
USA, Arkansas
BIG ROCK QUARRY
Pulaski County
Pulaski County
127
Geologic Map of Arkansas
N
Big Rock Quarry
Prepared by the Arkansas Geological Commission
and the United States Geological Survey, 1993
128
Jackfork Sandstone, Bigrock Quarry, Arkansas
Face-On view of outcrop
3D Photorealistic data for interpretation of deep
water facies.
Courtesy of Veritas
129
Jackfork Sandstone, Bigrock Quarry, Arkansas
Face-On view of outcrop
3D Photorealistic data for interpretation of deep
water facies.
Courtesy of Veritas
130
(No Transcript)
131
Big Rock Quarry
Channel
A. Three-dimensional model of the Big Rock Quarry
outcrop. B. Three-dimensional view of a channel
connecting the two sides of the quarry.
Most paleocurrents suggest a SW flow direction
few of them have SE markers. However in
most cases it was only possible to tell the
direction of the flow and not its sense. C.
Channel boundaries highlighted on the 3D model
allow reconstruction of submarine channel
complex architecture.
132
(No Transcript)
133
Digital mapping
GPR line
How photorealistic acquisition works!
  • Equipment used
  • Leica 500 RTK GPS system
  • MDL Laser Atlanta System
  • MDL Laser ACE
  • Topcon GPT 1002
  • Camera Fuji S1pro.

outcrop
Each photo and scan needs 5 control pts
134
Panther Tongue sandstone, Utah
UTAH
135
Processing Steps
3D model The 3-D model of the outcrop can be
digitized and the points representing bedding
surfaces will be in global coordinates. From the
bedding diagrams the surfaces can be reconstitute
in 3-D if the bedding contour is highly variable
in XYZ allowing a subsurface fit, or if
subsurface information such as
ground-penetrating radar data is available,
providing position of the contact seen in
exposure facilitating surface fitting of more
than one line in space.
136
Interpretation
Surfaces Using Gocad TM software bedding
surfaces topography and orientation of
paleochannels was accomplished. In GocadTM
channels features show relief of about 2-3
m. Channel features can be followed more than 20
m behind the outcrop and are oriented North
-South, identical with paleocurrent data from
cross strata and sole marks.
137
Ainsa Outcrops, Spain (with Norsk Hydro)
  • Ainsa Turbidite System
  • Four outcrops have been mapped
  • Main face 21 photos, about 1.2 km long and 70 m
    high
  • Canyon outcrop 9 photos, about 300 m long and 40
    m high
  • Quarry outcrop 5 photos, 250 m long and 20 m
    high
  • Road outcrop 1 photo, 90 m long, 35 m high
  • Most of photos have been shot from a helicopter,
    using 50 mm lens at 3040X2016 resolution
  • Models will be used for both training (virtual
    field trips) and for generating reservoir models
    of turbidite channel geometries.

Spain
138
Ainsa, Spain
  • Cooperative project with Norsk Hydro A/S
  • Ole Martinsen, Tore Loseth
  • Integrated data set Deep water siliciclastics
  • Photorealistic Models
  • Satellite/Aerial Imagery
  • Integration with other consortia data
  • Reservoir model constructed
  • Technique paper being prepared

139
Ainsa Location
after Cronin et al., 2002
140
Ainsa, Spain Virtual Model
141
Ainsa data integrated with available airborne
imagery and DEM
142
Applications of 3D Outcrop Data
  • John Thurmond

143
Immersive Environment (CAVE)
144
LASER SCANNERS
  • Imaging system provides a user with a dense set
    of three-dimensional vectors to unknown points
    relative to the scanner location
  • Unprecedented density of geospatial information
    coverage
  • Return-beam detection device
  • Beam deflection mechanism.
  • Controlled by laptop computer that is also used
    for data acquisition.
  • Range measurement is derived from the two-way
    travel time of a laser pulse
  • Orientation (elevation and azimuth) of the
    transmitted
  • Pulse is measured by the beam deflection system
  • Energy of the return pulse is also recorded
  • color (RGB) is also recorded

145
LASER SCANNER ACCURACY
  • BOEHLER, Vincent and Marbs, 2003.
  • Tested scanners for accuracy
  • Application was for cultural heritage
    applications
  • Manufacturers specs not comparable

146
Accuracy
  • Angular accuracy
  • Angles from combination of deflection of rotating
    mirrors and rotation about a mechanical axis
  • Provides position with range position
  • Range accuracy
  • Time of flight or phase comparison between
    outgoing and returning signal
  • Noise-fuzz of points on a flat surface

147
Resolution
  • Resolution
  • Userability to detect an object in point cloud
  • Two specs contribute
  • Smallest increment of angle between successive
    points (can manually set)
  • Size of laser spot
  • Edge effects
  • When a spot hits and edge and gets 2 locations
    and or 2 materials

148
Other effects on accuracy
  • Surface reflectivity
  • Distance, atmospheric, incidence angle
  • Albedo (ability to reflect)
  • White strong, black weak
  • Depends on spectra of the laser (green, red, near
    IR)
  • Shiny-poor reflector
  • Effects accuracy-range errors larger than specs

149
Environmental conditions
  • Temperature-check specs
  • Atmosphere-
  • changes propagation speed slightly
  • Dust, mist, fog-- a problem
  • Interfering radiation
  • Sunlight strong relative to signal
  • Influence or prevent (dont shoot into sun)

150
Other considerations
  • Measuring speed
  • Range limits
  • Field of view
  • Laser classeye safe?
  • Can register?
  • Can transform into coordinates?
  • Logistics-weight, batteries etc.

151
Measuring noise in range direction. Riegl Z420 is
comparable to Z360
152
Resolution test
153
Resolution test
154
Advantages and Disadvantages of mid-range laser
scanners
155
LASER SCANNERS(From POB magazine)
156
Other characteristics of previous
157
More scanners
158
(No Transcript)
159
(No Transcript)
160
(No Transcript)
161
SCANNER SOFTWARE
162
More on scanner software
163
Using fast laser scanners
  • Changed many of our procedures
  • Millions of points instead of thousands
  • Software limitations
  • Hardware limitations
  • BUT
  • Point clouds could integrate internally
  • Features such as vegetation definable
  • Scan at 1000to 24000 points/sec so much faster
  • Map at more detail
  • Can use the points in a cloud to position a
    location

164
Mineral Wells, Texas
  • Syn-sedimentary growth
  • (listric) faults
  • Pennsylvanian Mingus
  • Formation
  • The mapped outcrop is
  • about 260 m long and 30 m high
  • Camera original Fuji S1 Pro
  • (3040X2016 at 24 mm lens)
  • Next Canon D60 (6mb pixels)

165
Lecture Overview of Field Trip"Growth
Faulting and Depositional Environments of the
Pennsylvanian"
166
Mineral Well Outcrop
Note the detail yet the terrain interval is
several decimetersthe photo brings in millions
of pixels in detail.
167
Mineral Wells (Dobbs Valley)
  • Association of Engineering Geologists requested a
    virtual model be built of the Dobbs Valley growth
    fault outcrops.
  • Wanted to take 3D model of outcrops into field on
    field trip so they could fly around the outcrop
    while in the field.
  • UTD acquired data from across the Brazos River of
    dip and strike sections at 100-200 meters with
    LPM 3800 in one day.
  • GoCAD models of these integrated with Steeles
    geologic mapping.

168
Photos from AEG of outcrops from across river
Strike section
Dip section
169
Example of photomosaic interpretations from
McLinjoy.
170
Dobbs Valley Growth fault outcrop.Geology by
McLinjoy.Mapping of data onto 3D modelby Erik
Brandlin.
Left entire outcrop. Lower with McLinjoy data
171
Example of McLinjoy mapping color coded
(lines)and 3D analysis (surfaces)
172
Digitized McLinjoy data boundaries (lines) and
interpreted surfaces
Mapped features (lines) and analysis (surfacefits
173
North end of outcrop where original mapping was
done.
Xxx
Close up of north end of outcrop which was also
originally mapped with robotic laser.
174
Only mapped and interpreted geology shown.
175
(No Transcript)
176
This model used today by J. Bhattacharyas AAPG
field course
177
Mt. Rushmore, South Dakota
  • Mapped like tourists with big equipment
  • scanning from 3 locations in visitor area-two
    half days
  • Three photos over the 2 days
  • 10 million points, 2-3cm accuracy and resolution
    with LPM 3800 Riegl
  • Will return to add more data from top of heads

178
Entire modelmerged scans
179
Surface fit to scanned points
180
Final photorealistic model
181
Cretaceous Ferron Delta, Utah
182
GPR Equipment and Techniques
GPR Antenna
GPR Data Recorder
183
Cliff Face at Coyote Basin Conventional data from
5 years ago
184
Example of 3-D GPR Interpretation of GPRlayers
relate to geology
3D Relative Amplitude Display
185
LPM scanning of outcropLower leftintensity plot
of a scanLower right3D photorealistic model
Will see this again in stereo.
186
Using faster scanner at Salina tunnel and Mt.
Rushmore
Shorter Range scanner - 100 m - 24 KHz
- 1.2 cm Accuracy
187
Z360 in action in Salina tunnel.
188
Z360 in action
189
Salina Tunnel, Utah(note fault)
190
I-35 Arbuckle Outcrops
  • Mapping series of famous outcrops along I-35 in
    Oklahoma
  • Faulted anticlinein process
  • Unconformityshown here in Polyworks

191
I-35 unconformity model
Surface fit to point cloud
3D Virtual model
192
(No Transcript)
193
3D Photorealistic Modeling Process
194
Different Sensors
  • Scanners
  • Local coordinate system
  • Cameras
  • Local camera coordinate system
  • GPS
  • Global coordinate system

195
Coordinate Systems
  • Individual local scanner coordinates (each scan)
  • Object coordinate system (single coordinate
    system aligning all scans)
  • Camera coordinate system (each photograph)
  • Global coordinates

196
Scanner Coordinate
Z
  • Individual scanner local coordinate
  • Not necessary to level

Y
X
197
Camera Coordinate System
  • Each photograph has its own coordinates
  • Units mm or pixel

198
Putting it together
  • From individual scan coordinates to object
    coordinates
  • From object (or global) coordinates to camera
    coordinates
  • From object coordinates to global coordinates

199
Individual coordinates to object coordinates (1/2)
  • Traditional survey approaches
  • Need to level the scanner
  • set up backsight
  • Knowing scanner location and backsight angle
  • transform each point to the object coordinate
    system, usually global.
  • Advantage
  • easy to set up
  • one-step from local to global coordinates.
  • Disadvantage
  • problem in generating mesh models.

200
From individual coordinates to object coordinates
(2/2)
  • Use mesh alignment techniques (Polyworks)
  • No need to level.
  • Requires overlap with common features to minimize
    the distance.

Z
T
Y
sc1
sc2
X
201
From Object to Camera (1/2)
  • Two approaches
  • Polynomial fit (rubber sheeting)
  • Low accuracy,
  • No need to know camera intrinsic parameters
  • Projection transform (pinhole model)
  • High accuracy

202
From Object to Camera (2/2)
  • From object to camera coordinate system (pin hole
    model)
  • Perspective projection to convert to image
    coordinates (uv, pixel, or mm)

6 unknowns assuming known f Nonlinear-needs
initial value
203
Camera Calibration
  • Correct lens distortion
  • Radial distortion
  • Tangential distortion
  • Calculate f, k1, k2, p2, p2 in the lab for each
    lens.

204
Example of the calibration (Canon 17mm)
1
2
  • Radial distortion
  • Tangential distortion
  • Complete model

3
205
Example
Iteration 8 Residuals pts51 -0.0027
-0.0065 pts50 0.0045 0.0085 pts2034 0.0050
0.0087 pts 2010 -0.0066 -0.0100 omage0.08839
938218814 phi1.36816786714242 kappa
1.45634479894558 X -0.975 Y 0.519 Z -0.013
206
Bundle Adjustment
Adjust the bundle of light rays to fit each photo
207
Bundle Adjustment (2/2)
  • Photo no 7734
  • pt no U V
  • 201 -0.000 -0.000
  • 202 0.000 0.003
  • 203 -0.000 -0.006
  • 14 -0.003 0.012
  • 15 0.000 0.001
  • 204 0.001 -0.004
  • 205 0.001 -0.009
  • 302 0.001 0.004

Photo no 7735 pt no U
V 14 0.003 -0.006
15 0.003 -0.001 204
-0.001 0.004 205
-0.001 0.009 16 0.017
0.005 206 0.000
0.001 207 -0.001 -0.009
208 0.001 0.010 302
-0.006 -0.009
Photo no omega phi kappa X
Y Z 7733 3.5147 78.25411
85.03737 -1.031 0.628 0.046
7734 21.026 79.86519 68.09084 0.419
14.735 -1.055
208
From Object to Global (1/2)
  • 7-parameter conformal transformation

Where m11 cos(phi) cos(kappa) m12
-cos(phi) sin(kappa) m13 sin(phi) m21
cos(omega) sin(kappa) sin(omage) sin(phi)
cos(kappa) m22 cos(omage) cos(kappa)
sin(omega) sin(phi) sin(kappa) m23
-sin(omage) cos(phi) m31 sin(omage)
sin(kappa) cos(omage) sin(phi)
cos(kappa) m32 siin(omage) cos(kappa)
cos(omage) sin(phi) sin(kappa) m33
cos(omage) cos(phi) and s is scale factor
209
Transform to Global (2/2)
GPS
Object
Iteration5 scale 0.998986 () omega
0.22279535 phi -0.04740587 kappa
1.45393837 X trans 24.834 Y
trans 11.698 Z trans 2.142
Pt 1, X -0.012 Y 0.042 Z 0.010 Pt 2, X
0.008 Y -0.004 Z -0.012 Pt 3, X
0.011 Y 0.012 Z 0.004 Pt 4, X -0.017
Y -0.032 Z -0.007 Pt 5, X 0.010 Y
-0.018 Z 0.006
210
Surface Generation
  • Through merge process in Polyworks
  • Through fitting through GoCad
  • Through direct triangulation (Delauney
    triangulation, TIN)

211
Surface cleaning (in Polyworks)
  • The single most time consuming part of entire
    process (90 of time).
  • Filling the holes (because of scan shadow)
  • Correct triangles

212
Summarize
About PowerShow.com