CAPTURE AND GEOLOGICAL DATA
EXTRACTION: TOOLS FOR A BETTER
ANALYSIS AND DIGITAL MODELLING
UNIVERSITAT DE BARCELONA
U
BDavid García-Sellés
Pablo Granado
Oscar Gratacos
Núria Carrera
Pau Arbues
Digital Workflow
Outcrop
Digital Capture
Data Extraction
Inte
rpre
tati
on
A
na
lysis
F
ield
Wo
rk
Lidar
Photogrammetry
Million coordinates XYZ
Reflectivity
Texture color
Remote Sensing
Petrel, Move, Fracaflow Structure
Specific Modules
Rock slope
instability
DFN Digital
Fracture Network
Surfaces
Lineations
Digitalization
from 2D to 3D
Semi-automated
Supervised method
Specific Software
Capture Digital Outcrops
Lidar Equipment
Electric generator Laser scanner
Antenna
and DGPS
receiver
Digital camera calibrated
FIE
LD
WO
RK
Lidar Processing
Image 1 Image 2 Image 3
Requires logistical
High data quality
Total reliability
For advanced users in:
GPS differential, post-
processing Lidar data.
Añisclo anticline, Pyrenees. Paleocene
limestones
FIE
LD
WO
RK
GPS device
Photogrammetric SfM Equipment
Digital Camera
FIE
LD
WO
RK
Photogrammetric SfM Equipment
GPS positioning
attached to file picture
(exif).
Poor precision
must be
compensated
by a good
distribution of
photo-shooting
FIE
LD
WO
RK
Photogrammetric SfM Processing
Fast acquisition
Cameras position
distribution
Standard camera
Camera position Abra del Condor anticline, Bolivia. Sub Andean
Range. Sandstones
FIE
LD
WO
RK
Photogrammetric Model in Render view
Medium-high data
quality in accuracy
and density
Uncertain result
All users
FIE
LD
WO
RK
Abra del Condor anticline, Bolivia. Sub Andean
Range. Sandstones
Photogrammetric SfM Processing
Abra del Condor anticline, Bolivia. Subandean
Range. Graystones
FIE
LD
WO
RK
Photogrammetric Model in Photorealistic view
Medium-high data
quality in accuracy
and density
Uncertain result
All users
As a payload
Photogrammetric SfM Processing
Photogrammetry Vs Lidar
Differences in GPS devices: Scale Azimuth Dip Positioning
Lidar Model Photogrammetric Model
FIE
LD
WO
RK
Abra del Condor anticline, Bolivia. Sub Andean
Range. Sandstones
Photogrammetry Vs Lidar
Differences in GPS devices: Scale Azimuth Dip Positioning
FIE
LD
WO
RK
Abra del Condor anticline, Bolivia. Sub Andean
Range. Sandstones
Photogrammetric Vs Lidar
Offset corrected
FIE
LD
WO
RK
Abra del Condor anticline, Bolivia. Sub Andean
Range. Sandstones
Photogrammetric Vs Lidar
1 m
Comparative profile: Lidar Model – Photogrammetric Model
FIE
LD
WO
RK
Alignment histogram models
Abra del Condor anticline, Bolivia. Sub Andean
Range. Sandstones
Data Extraction from High Resolution Models
Surfaces
Simplified geometric
surfaces Point Cloud
Data extraction: Vectorizing point clouds algorithm I
NT
ER
PR
ETA
TIO
N
Stereoplot
Dip
- Points by regression
- Degree of fit (coplanarity, M)
- Reliability (colinearity, K)
Parameters obtained: - Vector orientation Azimuth
Point Cloud
Data extraction from surfaces: Planar regression by Moment of Inertia analysis I
NT
ER
PR
ETA
TIO
N
Dip
- Points by regression
- Degree of fit (coplanarity, M)
- Reliability (colinearity, K)
Parameters obtained: - Vector orientation Azimuth
Stereoplot
IN
TE
RP
RE
TA
TIO
N
0
2
4
6
8
10
12
14
16
18
20
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Mil
lio
n p
oin
ts
Time processing (hour)
Data extraction from surfaces: Outcrop vectorization
Coplanarity Index
0 6.2 x
Data extraction from surfaces: Planar regression by Moment of Inertia analysis
Coplanarity Index mask
0 6.2 x
Data extraction from surfaces: Planar regression by Moment of Inertia analysis
Coplanarity Index mask
0 6.2 x
Data extraction from surfaces: Planar regression by Moment of Inertia analysis
Colinearity Index
0 2 x
Data extraction from surfaces: Planar regression by Moment of Inertia analysis
Colinearity Index mask
0 2 x
Data extraction from surfaces: Planar regression by Moment of Inertia analysis
Colinearity Index mask
0 2 x
Data extraction from surfaces: Planar regression by Moment of Inertia analysis
Data extraction from surfaces: Outcrop vectorization
Outcrop section in detail
IN
TE
RP
RE
TA
TIO
N
Abra del Condor anticline, Bolivia. Sub Andean
Range. Sandstones
Data extraction from surfaces : Azimuth –Dip classification by sets
Fracture Sets
IN
TE
RP
RE
TA
TIO
N
Data extraction from surfaces : Isolate clusters-Reconstruct surfaces
Fracture Sets
IN
TE
RP
RE
TA
TIO
N
Data extraction from surfaces : Isolate patches-Reconstruct surfaces
Morphological Model
IN
TE
RP
RE
TA
TIO
N
Data extraction
Point cloud area
RGB Image
IN
TE
RP
RE
TA
TIO
N
High Resolution Model in render view
Geological feature as
geometric lineation or
geometric surface
High Resolution Model
Image outcrop
IN
TE
RP
RE
TA
TIO
N Data extraction from lineations: Image-Model link by Colinearity equation
Image coordinate Xi, Yi Model Coordinate: Xm, Ym, Zm Camera position
Cotiella Massif, Pyrenees. Cenomanian
limestones
Camera orientation
Focal length, lens distortion
IN
TE
RP
RE
TA
TIO
N Data extraction from lineations: Software implementation
Cotiella Massif, Pyrenees. Cenomanian
limestones
IN
TE
RP
RE
TA
TIO
N Data extraction from lineations: Software implementation
Cotiella Massif, Pyrenees. Cenomanian
limestones
Set 0
Points Selected by digitalizing
Planar regression calculated by the Inertial Moment
Azimut
Dip
- Number of used points
- Degree of fit (coplanarity, M) - Reliability (colinearity, K)
Parameters obtained:
- Vector orientation
- Trace length
Stereoplot
IN
TE
RP
RE
TA
TIO
N Data extraction from lineations: Software implementation
Cotiella Massif, Pyrenees. Cenomanian
limestones
Image coordinate Xi, Yi Model Coordinate: Xm, Ym, Zm
Selected points respect to line distance tolerance in order to model density
IN
TE
RP
RE
TA
TIO
N Data extraction from lineations: Image advantages
Cotiella Massif, Pyrenees. Cenomanian
limestones
IN
TE
RP
RE
TA
TIO
N Data extraction from lineations: IR image
Cotiella Massif, Pyrenees. Cenomanian
limestones
IN
TE
RP
RE
TA
TIO
N Data extraction from lineations: Photogrammetric solutions, orthoimages
Cotiella Massif, Pyrenees. Cenomanian
limestones
Digitalization at field, before acquisition
or at desktop, after processing High Density Model
Analysis
Surface Detection: Isolate patches-Reconstruct surfaces
Plane Orientation
Centroid
Points used
Rughness
Approachs in
Wide
Long
Area
Morphological model
IN
TE
RP
RE
TA
TIO
N
Surfaces
Lineations
Plane Reconstruction
Morphological Model
Analysis: Morphological model for analysis
0123456789
10111213141516
Number of fractures/m
Number of fractures/m Scanline data:
Spacing Mean
Number of fractures / Scanline
Density of fractures
Orientation Mean
Frequency: Area, spacing....
AN
ALY
SIS
Analysis: Morphological model
Histograma of logarithmic base for plane area and set
AN
ALY
SIS