20-04-23

Challenge the future

DelftUniversity ofTechnology

Martijn Meijers

STW User Committee meeting, Utrecht, 19 September, 2012

Vario-scale (and nD extension): results and work in progress

2Vario-scale

Contents

• Creating smooth SSC• Using SSC• 3D vario-scale• Conclusion

tGAP = topological Generalized Area Partitioning (2D)SSC = Space Scale Cube

3Vario-scale

Smooth simplify

•Shock change:• 2 rectangles• 1 triangle

•Smooth change:• 3 triangles

Original line

Simplified

Transition

4Vario-scale

Smooth merge for convex neighbour

1. make #nodes shared top and target bnd equal (n) view

2. connect node pairs

3. 2 triangles + n-3 quadrangles

4. if non-flat space-split quadrangle scaleinto 2 triangle view

5. Merge planar neighbours

5Vario-scale

Alternative (without adding nodes)

LS_s = 7

LS_o = 1+3+1 = 5

b_s (shared boundary)

b_o (opposite boundary)

at scale s2, from s1 to s2, at scale s1

Create triangle base at one sideand top at other side

6Vario-scale

Non-convex neighbour subdivide in convex parts

m-shaped neighbour neighbour with hole

(note: smooth collapse/split similar to smooth merge)

7Vario-scale

Implementation: simplification

8Vario-scale

Contents

• Creating smooth SSC• Using SSC• 3D vario-scale• Conclusion

tGAP = topological Generalized Area Partitioning (2D)SSC = Space Scale Cube

9Vario-scale

Selection based on (n+1)D overlap from space-scale cube

Simple initial map Progressive initial map(sorting lowerhigher detail)

x y

s

High LoD

Low LoD

10Vario-scale

Selection based on (n+1)D overlap from space-scale cube

Simple initial map Progressive initial map(sorting lowerhigher detail)

x y

s

11Vario-scale

(n+1)D overlap selection for zooming

Progressive zoom-in Progressive zoom-out

(normal sorting order) (reverse sorting order)

12Vario-scale

(n+1)D overlap selection for panning

Normal panning Progressive panning

(normal sorting order)

13Vario-scale

tGAP data

WFS client

HTTP Get/ Post

jdbc: sql query

Progressive rendering

Web server

GML stream

OGC Web Feature Service (WFS)

Display

Google Earth (client)

HTTP Get

Multi- scale rendering

Web server

Display

Python/ GDAL application

oci: sql result set (polygons)

oci: sql query

Oracle Spatial

KML slice

jdbc: sql result set (polygons/ faces, BLG- trees and nodes)

tGAP data

WFS client

HTTP Get/ Post

jdbc: sql query

Progressive rendering

Web server

GML stream

OGC Web Feature Service (WFS)

Display

Google Earth (client)

HTTP Get

Multi- scale rendering

Web server

Display

Python/ GDAL application

oci: sql result set (polygons)

oci: sql query

Oracle Spatial

KML slice

jdbc: sql result set (polygons/ faces, BLG- trees and nodes)

14Vario-scale

Streaming of importance range(and first compared with a cut)

•A cut (or slice) of single importance

•A ordered range of importance values

select face_id as id, '101' as impLevel, RETURN_POLYGON(face_id, 101) as geom from tgap_face where imp_low <= 101 and 101 < imp_high;

select face_id as id, imp_high as impLevel, imp_low, imp_high, RETURN_POLYGON(face_id,imp_high) as geomfrom tgap_face where imp_high >= imp_high order by imp_high desc;

15Vario-scale

Extension to OGC/ISO WFS

• It is possible to specify imp range in Filter part of GetFeature request and using ogc:SortBy

• Not ideal because it is not clear that this is about scale, streaming, progressive transfer/refinement

• Deeper integration in WFS (called WFS-Refinement):1. GetCapabilities should indicate if server supports

progressive refinement2. Reporting of the min and max imp of a theme3. New request type GetFeatureByImportance

16Vario-scale

Example WFS-R request

<wfs:GetFeatureByImportance service="WFS" version="1.0.0" outputFormat="GML2" ...> <wfs:Query typeName='tgap_face' minImp='5' maxImp=‘8'> <ogc:Filter> <ogc:BBOX> <ogc:PropertyName>geom</ogc:PropertyName> <gml:Box srsName=“…epsg.xml#28992"> <gml:coordinates> 136931,416574 139382,418904 </gml:coordinates> </gml:Box> </ogc:BBOX> </ogc:Filter> <ogc:SortBy>gdmc:imp_high D</ogc:SortBy> </wfs:Query></wfs:GetFeatureByImportance>

17Vario-scale

Contents

• Creating smooth SSC• Using SSC• 3D vario-scale• Conclusion

tGAP = topological Generalized Area Partitioning (2D)SSC = Space Scale Cube

18Vario-scale

3D smooth merge

zizii

zi’’zii’’

zi’zii’

No ZI

ZII

19Vario-scale

Generic 3D smooth mergeHigh detail

zi’

zii’

1 (bottom)

3 (top)

4 (back)

1

2 (front) 2

45

3

5 (left)

c

b

da

C

B

A

D

e

f

5 (left) zi zii

1 (bottom)

3 (top)

4 (back)

2 (front)

Opposite boundary (5 faces)

Equator (ring)

zi’’

zii’’

Low detail

20Vario-scale

3D Smooth simplify

ZII’’

ZII’

ZII

Simplify boundary of merged object, two options:

1. Keep block shape

2. Tilted roof shape

21Vario-scale

Pseudo 4D-view

Zone zi shown for reference purpose

x

y

z

s

zi zii ZII’’

s1 s2

22Vario-scale

Step-wise approach

23Vario-scale

Contents

• Creating SSC• Using SSC• 3D vario-scale• Conclusion

tGAP = topological Generalized Area Partitioning (2D)SSC = Space Scale Cube

24Vario-scale

Conclusion, advantages vario-scale maps

•Map producers: more integrity, less manpower needed for the production of maps (=less cost)

•Map providers: more efficiency, less data = less storage capacity, less energy, high transmission speed

•Map users: more functionality at less cost and higher speed

•Patent pending nr. OCNL 2006630

25Vario-scale

Conclusions, true vario-scale

• True vario-scale nD maps based on (n+1)D representations and slicing (selecting) with hyperplanes:

• tGAP structure translates 2D space and 1D scale in an integrated 3D topological representation: no overlaps and no gaps (in space and scale)

• Starting with 3D space and adding scale results in 4D• Starting with 3D space and time (history) and adding

scale results in 5D topological structure (again no gaps/overlaps in space, time or scale), well defined neighbors in space, time and scale directions