Challenges in Photogrammetryand Remote Sensing
Norbert Pfeifer
Vienna University of Technology
Institute of Photogrammetry and Remote Sensing I.P.F.
Statements
„Photogrammetry is finished“Professor Rinner (bundle block formulation) 1960ies
„Remote Sensing has always been the method of the future“Saying in Forest Science
Old stuff, that never will work ?
2
Economic Predictions
Mapping Opportunities, Gewin, 2004: Nature 427(6972)21st Century growth markets seen by US Bureau of Labour• Nano-technology• Bio-technology• Geo-technology
The economic value of the Dutch geo-information sector, Castelein et al., 2010: Int.J. Spatial Data Infrastructure Research Geo-Information sector is 0.25% of Dutch GDP
3
Obvious Demands
Geoinformation for Disaster and Risk Management, Altan et al. (eds.), 2010: Presented at UNOOSA July 2010.
4
Obvious Demands
Geoinformation for Disaster and Risk Management, Altan et al. (eds.), 2010: Presented at UNOOSA July, 2010.Climate Change and Adaptation• Earth Observation • Monitoring• Remote Sensing
5
Obvious Demands
Resources• Forests is 25% of land mass• Fauna-Flora-Habitat Directive• Megacities, space consumption• Cultural
Heritage
6
natura2000.eea.europa.eu
Afghan Buddha Statues, TU Wien, Faculty of Architecture and Spatial Planning
to moving measurement platformsfrom single to multi-sensor systemsto higher autonomy and automationto new fields of applications
Presentation outline
the art and science of infering metric information from imagesPhotogrammetry, Remote Sensingand their neighboring disciplines
ChallengesSensors, Methods, and Applications
Photogrammetry
9
Geo…Geometry the Earth
object-based area-based
Germanisches Nationalmuseum Nürnberg
Photogrammetry
10
Geo…Geometry the Earth
object-based area-basedImage understanding
PhysicsProcesses
object based area based
Photo-grammetry
from States to Processes
11
P & RS neighbor: Spatial Information Science
Geo-data infrastructures (GDI) and ubiquituos geo-informationVirtual Globes: NASA, Google, Microsoft, …
Navigation LBSWeb 2.0 geodata mash-ups
Omnipresent Geo-Data
Omnipresent Geo-data
Geo-data infrastructures (GDI) and ubiquituos geo-informationVirtual Globes: NASA, Google, Microsoft, …
Navigation LBSWeb 2.0 geodata mash-ups
Wiki mapping, collaborative mapping, crowd sourcing, …
• interest driven• update stream
Free global Geo-data• SRTM• CBERS, … open street map
Omnipresent Geo-data
Methods for collaboration not yet settledAcademic contribution to Open Source GDI very little
Free national Orthophotos, US-Lidar for the nationEuroSDR: Crowd sourcing for updating national databases
Commercial GDI „Open-Source“ GDI
National GDI
P&RS Mission
geo-data and modelsfor a sustainable development
of the natural and cultural environment
Exploit new sensor technologyIncrease automation in modelingAdopt new applicationsStrengthen Sensor – Method – Application feedback
Where‘s the challenge ?
geo-referencedand geo-physical
New Sensors
Vienna, July 4, 2010 17
Vexcel Patent
SPOT, Microsoft, Leica, Riegl
Standard Imaging Sensors
Texture sensors• Digital aerial camera: digital workflow• Digital semi-metric and consumer cameras• Hi-Res/Med-Res RBGNir Satellite Imager (hardly large image blocks!)
Radiometric and spectral sensors• Multispectral Satellite Imager• Airborne Imaging Spectroscopy• Microwave/SAR Satellite Imager• FLIR cameras
Range sensors• Aerial/Terrestrial Laser Scanner• Active Triangulation• Structured Light
Leica, Specim, Optech, Minolta, SMOS
After Gordon Petrie
Sensors Developments
Sensors combinationsRadiometric exploitation of digital aerial cameras (EuroSDR-Test, FGI)Full Waveform Laser ScanningMulti-spectral Laser ScanningTime-of-Flight CamerasSingle Lens Stepping FrameTerrestrial InSAR
Antonov of Geokosmos: Optech, Z/I, Rollei
© xsens, MEMS IMU
© u-blox GPS/Galileo chip
Challenges
Tighter sensor integration physically
Integrated processing of data streamse.g. of GNSS, INS, LiDAR, and imaging spectrometer
Corresponding features, orientation
Calibration: geometrically and geo-physically
Understanding measurement processes
Approach
Tigther integration of physical and geometrical views
Coupling of Orientation, calibration and primitive modeling
Improve Automation
Photogrammetry: Lack of AutomationRemote Sensing: Lack of Reliability / Transferability
21
120 km2
SPOT Classification by Kressler, Kim, Steinocher
148,940,000 km2 Land
Improve Automation
Photogrammetry: Lack of AutomationRemote Sensing: Lack of Reliability / Transferability
22
120 km2
SPOT Classification by Kressler, Kim, Steinocher
148,940,000 km2 Land
Efficientrepresenation and retrieval
of implicit and explicitknowledge and experience
Standardized Methods and Products
Satellite image classifications
Surface/terrain models, orthophotos, point clouds
3D city models (roof landscape, facades)
Challenges
Modeling from different sources: time, scale, observableReliability, correlated observations, quality controlRepresenation and retrieval of knowledge and experienceContradictory informationProcessing time
ApproachExploit redundancy, compressionAdvance from empirical to geo-physical process modelsExploit machine learning competence
What is a house ?Difference: mirror left – mirror right
Different sources example …
Forest inventory data and airborne laser scanningAirborne and terrestrial point clouds
© Doneus, Briese, Studnicka
W. Dorigo, M. Hollaus, W. Wagner, K. Schadauer: An application-oriented automated approach for co-registration of forest inventory and airborne laser scanning data; Int. J. of Remote Sensing, 31 (2010), 5; 1133 - 1153.
New Applications
Cartography, Urban Planning, GDIIndustry, Cultural Heritage, Hydrology
Quantification ProcessesGeography, Geomorphology, GeologyBiology, Ecology, Forestry
Vienna, July 4, 2010 Beyond the ISPRS Centenary 26
Simon Buckley, Norway
biodiversity, hydraulic resistance, timber volume, channel erosion, dead ice, grain size distribution, …
Sensors – Methods – Applications
Feedback – Feed Forward – Loops
27
… be scientifically sustainable bypersistent development of methods …
Loads of interesting work & research aheadOpenness for applications & sensors essentialModeling of geo-physical (and geo-social!) processesIntegration of geometry and physics3D model → global dynamic geo-physical etc. etc. model