aligning geospatial sources · introduction geospatial data sources have become widely available...
TRANSCRIPT
![Page 1: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/1.jpg)
Aligning Geospatial Sources
Ching-Chien (Jason) Chen
Geosemble Technologies http://geosemble.com
Prof. Craig A. KnoblockProf. Cyrus Shahabi
![Page 2: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/2.jpg)
OutlineIntroduction & MotivationOur approach: AMS-conflation
OverviewVector and imagery conflation
TechniquesExperimental results
Map and imagery conflationTechniquesExperimental results
DemonstrationsRelated WorkConclusion
![Page 3: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/3.jpg)
IntroductionGeospatial data sources have become widely availableAutomatically and accurately integrating two spatial datasets is a challenging problem
Focus on resolving spatial inconsistencies (positional inconsistencies) to align datasets
Orthoimagery( in raster format )
Street maps ( in raster format )
Road network( in vector format )
![Page 4: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/4.jpg)
Lat / Long
Lat / LongLat / Long
Lat / Long
ChallengesDifferent projections, accuracy levels, resolutionsresult in spatial inconsistencies
Name: Stanley SmithAddress: 125, Gabriel Dr.City: St. LouisState: MOPhone: (314)955-4200Lat: 38.5943572Long: -90.4265843
Road Name : Gabriel DrRange: 20 – 500Zip: 63103
Motivation :Vector and Imagery Integration
![Page 5: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/5.jpg)
Motivation :Vector and Imagery Integration in Google Earth
D.C.
Ft. Campbell, KY (only zoom into 1m/p)
![Page 6: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/6.jpg)
Motivation: Parcel Vector and Imagery Int. in Zillow
Zillow.com
![Page 7: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/7.jpg)
Lat / Long
Lat / Long
Lat / Long
Lat / Long
Integration ChallengeDifferent geographic projections and accuracy levels
Motivation :Map and Imagery Integration
![Page 8: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/8.jpg)
Another Integration ChallengeSome online maps are not
geo-referenced
Lat / Long
Lat / Long
?
?
Motivation :Map and Imagery Integration
![Page 9: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/9.jpg)
Motivation: No Map and Imagery Int. in Google Earth
Google’s hybrid view of map and imagery(like vector-imagery integration)
Our approach:Map-Imagery integration (MapQuest map)
Transportation Raster Map (showing Metro-Links and streets)
Population Density Map (showing Population Density and Streets)
![Page 10: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/10.jpg)
Traditionally, the problems of vector-imagery and map-imagery alignment have been in the domain of GIS and Computer VisionIn GIS literature
The alignments were previously performed manuallyCommercial products: MapMerger ; Able R2V; Intergraph I/RASC; ESRI ArcGIS
In Computer Vision literatureThe alignments were performed automatically based on image processing techniques
Often required significant CPU time
Motivation: The Status of the Art
![Page 11: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/11.jpg)
OutlineIntroduction & MotivationOur approach: AMS-conflation
OverviewVector and imagery conflation
TechniquesExperimental results
Map and imagery conflationTechniquesExperimental results
DemonstrationsRelated WorkConclusion
![Page 12: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/12.jpg)
Aligning Geospatial Data Using Conflation TechniqueConflation: Compiling two geo-spatial datasets by establishing the correspondence between the matched entities (control point pairs)and transforming other objects accordingly
Imagery
FindControl Points
Aligning the Vector and ImageryVector
![Page 13: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/13.jpg)
Our Approach: AMS-ConflationAutomatic Multi-Source Conflation
Exploiting multiple sources of geospatial information to automatically align data
Automatically exploiting information from each of the sources to be integrated to generate accurate control point pairsExploited geospatial information from one data source can help the processing of the other source
![Page 14: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/14.jpg)
Inferred information from the data source
Detected edgesDetected intersections
by corner detector• Connectivity: 3
• Orientations:10, 900, 1800
Detected edges Classified road pixels
Inferred information from the data source
Inferred information from the data source
Road IntersectionsRoad Directions
AMS-Conflation :Exploit Inferred Information from the Data Source
![Page 15: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/15.jpg)
Metadata about the data source Resolution
(or map scale)
153 m
Geo-coordinatesResolution
Long: -90.43Lat: 38.595
Long: -90.42Lat: 38.594
42 m
10 m
Metadata about the data source
Road widths
Metadata about the data source
AMS-Conflation :Exploit Metadata about the Data Source
![Page 16: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/16.jpg)
AMS-Conflation :Exploit Peripheral Datasets to the Data Source
![Page 17: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/17.jpg)
OutlineIntroduction & MotivationOur approach: AMS-conflation
OverviewVector and imagery conflation
TechniquesExperimental results
Map and imagery conflationTechniquesExperimental results
DemonstrationsRelated WorkConclusion
![Page 18: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/18.jpg)
AMS-Conflation to Align Vector and Imagery
Lat / Long
Lat / Long
Lat / LongLat / Long
Control Point Detection
Intermediate control points
Filtering Technique
Finalcontrol points
Triangulation and Rubber-Sheeting
![Page 19: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/19.jpg)
Aligning Vector and Imagery:Finding Control Point Pairs Using Localized Template Matching (LTM)
0
0.05
0.1
0.15
0.2
0.25
Hue (degree)
On-road
Off-road
BayesClassifier
Matching byCorrelation
road width and road directions
![Page 20: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/20.jpg)
Vector median
k% control-point vector
X (meter)
Y (meter)
-10
-10
10
10
O
W
Aligning Vector and Imagery:Filtering Control Point Pairs Using Vector Median Filter (VMF)
![Page 21: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/21.jpg)
Conflating Imagery and Vector Data Using Rubber-Sheeting [Saalfeld’88]
![Page 22: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/22.jpg)
- Red lines: Original MO-DOT - Blue lines: Conflated lines
Conflation Results:MO-DOT+ High-resolution USGS Color Image
![Page 23: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/23.jpg)
- Red lines: Original NAVSTREETS roads - Yellow lines: Conflated lines
Conflation Results:NAVSTREETS+ High-resolution USGS Color Image
![Page 24: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/24.jpg)
Evaluation
- Red lines: Reference roads (roadsides)
- Blue lines: Reference roads (centerlines)
Completeness : the percentage of the reference roads for which we generated conflated lines
(Length of matched reference roads)/(Length of reference roads)
Correctness : the percentage of correctly conflated lines with respect to the total conflated lines
(Length of matched conflated lines)/(Total length of conflated lines)
Positional Accuracy : the percentage of conflated roads within x meters to the reference roads
Using road-buffermethod
x xx xReference roads
Buffer zone of buffer width x
Conflated roads x xx xx xx xReference roads
Buffer zone ofbuffer width x
Conflated roads
![Page 25: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/25.jpg)
Results:US Census TIGER/Lines roads+ B/W Image
Original TIGER/Lines
Conflated TIGER/Lines
Completeness 37.9% 84.7%Correctness 31.3% 88.49%
PositionalAccuracy
Yellow Lines: Conflated TIGER/LinesRed Lines: Original TIGER/Lines
![Page 26: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/26.jpg)
Original NAVSTREETS Conflated NAVSTREETS
Completeness 44.9 % 74.4 %Correctness 47.9 % 85 %
PositionalAccuracy
Results:NAVTEQ NAVSTREETS+ High-res Image
Blue Lines: Conflated NAVTEQ NAVSTREETSRed Lines: Original NAVTEQ NAVSTREETS
![Page 27: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/27.jpg)
Aligning Parcel Vector Data with ImageryBlue Lines:
Conflated Parcel Vector Data
Red Lines: Original Parcel Vector Data
![Page 28: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/28.jpg)
OutlineIntroduction & MotivationOur approach: AMS-conflation
OverviewVector and imagery conflation
TechniquesExperimental results
Map and imagery conflationTechniquesExperimental results
DemonstrationsRelated WorkConclusion
![Page 29: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/29.jpg)
AMS-Conflation to Align Maps and Imagery
Detect IntersectionPoints On the Map
Map with Unknown Coordinates
Geo-referenced Imagery
?
?
Lat/Long
Lat/Long
Points On the Imagery/ Vector Data
Vector-ImageryConflation
Point PatternMatching
& Map-ImageryConflation
![Page 30: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/30.jpg)
Finding Intersection Points on MapsDifficult to identify intersection points automatically and accurately
Varying thickness of linesSingle-line map v.s. double-line mapNoisy information: symbols and alphanumeric characters
Proposed a technique to detect intersections in [acm-gis’04]Our primitive technique is further improved in [Chiang et al. acm-gis’05 ]
Maps
Find Corners(using OpenCV)
Apply morphological operator
Points
LinesRecognize intersections by checking line connectivityDetected
intersections
Remove noisy information by separating lines and characters
Isolate map data by automatic thresholdingand trace parallel lines
![Page 31: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/31.jpg)
Finding Intersection Points on Maps: Results
Identify Intersections
Some noisy points will be detected as intersection points. Our geo-spatial point matching algorithm can tolerate the existence of misidentified intersection points.
![Page 32: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/32.jpg)
Point Pattern Matching: Overview
Example: (x,y) = (83,22) Example: (lon,lat) = (-118.407088,33.92993)80 points 400 points
Distribution of feature points is the fingerprint of the spatial datasetFind the mapping between these points to get a set of control point pairsHow to solve the point sets matching problem :
A geometric point sets matching problemFind the transformation T between the layout (with relative distances) of the two point sets
![Page 33: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/33.jpg)
Point Pattern Matching:Finding the Transformation
Scaling in West-East direction
Scaling in North-South direction
Transformation = Scaling + Translation Transforms most points on map to points on
imageryFind matching point pairs to solve this
transformation
![Page 34: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/34.jpg)
Point Pattern MatchingPick a pair of points in first dataset
![Page 35: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/35.jpg)
Point Pattern MatchingPick two random points in the second dataset as potential matching candidates
![Page 36: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/36.jpg)
Point Pattern MatchingApply the same transformation to the other pointsRepeat this for a different pair until enough points are matched.
![Page 37: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/37.jpg)
Geospatial Point Pattern Matching (GeoPPM):Exploit geospatial properties to reduce the search space of PPM
Connectivity of an intersection: the number of intersected roadsOrientations of intersected road segmentsMap scalePoint density and localized distribution of the points
Connectivity:3; Orientations: 0, 90, 270
![Page 38: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/38.jpg)
GeoPPM: Exploit Connectivity
Impossible candidates for P1
P1
Possible candidates for P1
![Page 39: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/39.jpg)
GeoPPM: Exploit Orientation
Possible candidates for P1Impossible candidates for P1
P1
![Page 40: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/40.jpg)
GeoPPM: Exploit Angles between Points
Some of the possible Some of the impossible Candidate pairs for P1,P2Candidate pairs for P1,P2
P1
P2
![Page 41: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/41.jpg)
GeoPPM: Exploit Density between Points
Some of the possible Some of the impossible Candidate pairs for P1,P2Candidate pairs for P1,P2
P1
P2
![Page 42: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/42.jpg)
GeoPPM: Localized Distribution of the Points
![Page 43: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/43.jpg)
GeoPPM: Localized Distribution of the Points
![Page 44: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/44.jpg)
GeoPPM: Localized Distribution of the Points
![Page 45: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/45.jpg)
Using matched point pattern to align maps with imagery by rubber-sheeting
Aligning Maps and Imagery
![Page 46: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/46.jpg)
Using matched point pattern to align maps with imagery by rubber-sheeting
Aligning Maps and Imagery
![Page 47: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/47.jpg)
Experimental Setup:Two Areas with various Maps/Imagery
![Page 48: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/48.jpg)
Results
![Page 49: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/49.jpg)
The conflated map roads v.s.the corresponding roads in the imagery
Use TIGER maps for evaluationCompare conflated map roads with reference roads(manually plotted roads)
Completeness/ Correctness /Positional Accuracy
Reference roads
Conflated map roads
Evaluation:
![Page 50: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/50.jpg)
CompletenessCorrectness
010203040506070
%
Original TIGER mapConflated TIGER map
Completeness/Correctness Positional Accuracy
Evaluation:The Performance of Map-Imagery Conflation
![Page 51: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/51.jpg)
OutlineIntroduction & MotivationOur approach: AMS-conflation
OverviewVector and imagery conflation
TechniquesExperimental results
Map and imagery conflationTechniquesExperimental results
DemonstrationsRelated WorkConclusion
![Page 52: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/52.jpg)
DemonstrationsDemos using Heracles as the UI
Model vector-imagery conflation and map-imagery conflation as web services
Demos using browser as the UI (only for pre-picked areas and only for map-imagery conflation)
http://yoyoi.info/dcbus.htmlhttp://yoyoi.info/dcmapquest.html
Conflation.Web services
LocationImagery ConflatedResultsVector (or Map)
Http/SOAP
Http/SOAP
Client(Heracles)
Some knowledge about the vector and imagery (including learned road color)
![Page 53: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/53.jpg)
OutlineIntroduction & MotivationOur approach: AMS-conflation
OverviewVector and imagery conflation
TechniquesExperimental results
Map and imagery conflationTechniquesExperimental results
DemonstrationsRelated WorkConclusion
![Page 54: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/54.jpg)
Related WorkVector to vector conflation based on corresponding features identified from both vector datasets (in GIS domain)
[Walter et al. 99]: Matching features (e.g. intersection points or polygons) at geometry level[Cobb et al. 98]: Matching features both at spatial/non-spatial level[Chen et al. 06]: Matching two road networks using GeoPPM
Vector to imagery conflationUtilizing matched polygons [Hild et al. 98]Utilizing matched lines [Filin et al. 00]Utilizing matched junction-points [Flavie et al. 00]All above solutions
Require lots of CPU timeUtilize vector data only for verifying detected features not for extracting features
![Page 55: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/55.jpg)
Related WorkRaster to raster conflation:
To the best of our knowledge, there is no research addressing the problem of automatic conflation of maps and imageryRelated work of imagery-imagery conflation
Sato et al. [Sato 01] proposed an edge detection process was used to determine a set of features that can be used to conflate two image data sets
Their work requires that the coordinates of both image data sets be knownDare et al. [Dare 00] proposed multiple feature extraction and matching techniques
Need to manually select some initial control pointsSeedahmed et al. [Seedahmed 02] proposed an approach extract features from imagery by Moravec feature detector and obtain transformation parameters by investigating the strongest clusters in the parameter space
Require lots of CPU timeCommercial products: Able R2V and Intergraph I/RASC
Need to manually select all control points
![Page 56: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/56.jpg)
OutlineIntroduction & MotivationOur approach: AMS-conflation
OverviewVector and imagery conflation
TechniquesExperimental results
Map and imagery conflationTechniquesExperimental results
DemonstrationsRelated WorkConclusion
![Page 57: Aligning Geospatial Sources · Introduction Geospatial data sources have become widely available Automatically and accurately integrating two spatial datasets is a challenging problem](https://reader034.vdocuments.net/reader034/viewer/2022050510/5f9ae702517c896731721ad1/html5/thumbnails/57.jpg)
ConclusionAMS-Conflation
Automatic Vector to Imagery ConflationRoad Vector to Black-White Imagery Alignment [sstd’03]Road Vector to Color (High-resolution) Imagery Alignment [stdbm’04] [GeoInformatica’06 ]Parcel Vector to Color (High-resolution) Imagery Alignment [sstd’07 ?]
Automatic Map to Imagery Conflation [ng2i’03a] [acm-gis’04] [acm-gis’05] [GeoInformatica’07 ?]Automatic Vector to Vector Conflation [sstdbm’06]
Future Work:Improve and generalize AMS-conflation
Elevation data and hydrographic dataRelax the constraint of utilizing vector as glue to align maps and imageryContinuous efforts to commercialize AMS-Conflation