Efficient derivation and caricature of urban settlement boundaries - - PowerPoint PPT Presentation

efficient derivation and caricature of urban settlement
SMART_READER_LITE
LIVE PREVIEW

Efficient derivation and caricature of urban settlement boundaries - - PowerPoint PPT Presentation

Efficient derivation and caricature of urban settlement boundaries for 1:250k Andreas Reimer and Christian Kempf Universit at Heidelberg 1 / 28 Can schematisation techniques be used for medium scale generalisation? 2 / 28 Huh?


slide-1
SLIDE 1

Efficient derivation and caricature of urban settlement boundaries for 1:250k

Andreas Reimer and Christian Kempf

Universit¨ at Heidelberg

1 / 28

slide-2
SLIDE 2

Can schematisation techniques be used for medium scale generalisation?

2 / 28

slide-3
SLIDE 3

Huh? Schematisation?

3 / 28

slide-4
SLIDE 4

Huh? Schematisation?

Definition

We define schematisation in cartography as a process that uses cartographic generalisation operators in such a way as to produce diagrams

  • f a lower graphical complexity compared to maps of the same scale; the

process aims to maximise task-adequacy while minimizing non-functional

  • detail. In contrast, traditional cartographic generalization can be

understood as trying to maximise functional detail with task-adequacy (in the form of legibility) as a constraint. Schematisations use many unorthodox design principles → carricature etc.

3 / 28

slide-5
SLIDE 5

Why Strategi?

250k legacy product manually updated no relation to OS Master Map data geometrically fitted to generalised base data

4 / 28

slide-6
SLIDE 6

Why Strategi?

5 / 28

slide-7
SLIDE 7

Why Strategi?

6 / 28

slide-8
SLIDE 8

Approach

find out target design rules/constraints empirically reduce input complexity in model generalisation redraw geometries in cartographic generalisation

7 / 28

slide-9
SLIDE 9

Observations...

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 number of nodes/perimeter [linelength mm] 500 1000 1500 2000 2500 quantity

Strategi

8 / 28

slide-10
SLIDE 10

Input from OS Master Map

9 / 28

slide-11
SLIDE 11

Selection

10 / 28

slide-12
SLIDE 12

Buffer

11 / 28

slide-13
SLIDE 13

Union

12 / 28

slide-14
SLIDE 14

Selection by size

13 / 28

slide-15
SLIDE 15

Simplification: Visvalingam-Whyatt

14 / 28

slide-16
SLIDE 16

Angular schematisation

15 / 28

slide-17
SLIDE 17

Input and result

16 / 28

slide-18
SLIDE 18

Result in context

17 / 28

slide-19
SLIDE 19

At viewing scale

18 / 28

slide-20
SLIDE 20

Original Strategi

  • #
  • !
! ! ! ! ! ! !
  • RIVER

MERSEY RIVER DEE / AFON DYFRDWY Delamere Forest Ince Banks WIRRAL Gayton Sands White Sands Dungeon Banks Eastham Sands Dawpool Bank Bagillt Bank Sealand Sandycroft Mancot Royal Bretton Broughton Shotton Hawarden Drury Ewloe Northop Northop Hall Weston Hale Eastham Bromborough Raby Thornton Hough Barnston Gayton Flint Mountain Soughton/Sychdyn Hale Bank Grassendale Port Sunlight Storeton Pensby Irby Thurstaston Queensferry New Brighton Wepre Saltney Parkgate Ness Burton Puddington Shotwick Saughall Mollington Capenhurst Ledsham Willaston Hooton Childer Thornton Backford Stoak Wervin Picton Bridge Trafford Dunham-on-the-Hill Thornton-le-Moors Hapsford Alvanley Elton Ince Manley Mouldsworth Kingsley Newton Helsby Ashton Hayes Oscroft Duddon Burton Clotton Utkinton Willington Corner Kelsall Upton Mickle Trafford Great Barrow Tarvin Rowton Waverton Christleton Handbridge Littleton Guilden Sutton Blacon Lache Vicarscross Quarrybank Connah's Quay Buckley Speke Bebington Flint Garston Heswall Runcorn Neston Frodsham

176 158 14 10 8 7 4 15

A5032 A549 A56 A55(T) A5104 A540 A548 A5032 A540 A5119 A533 A551 A5480 A5137 A5117 A540 A548 A5117 A550 A41 A5119 A5032 B5463 B5439 B5132 B5125 B5132 B5130 B5125 B5132 B5132 B5153 B5151 B5137 B5134 B5132 B5129 B5155 B5393 B5137 B5128 B5136 B5132 B5171 B5125 B5132 B5152 B5151 B5128 B5133 B5135 B5126 B5136 B5138 B5463 B5151 B5155 B5441 M56 M56 M53 M53 M56 M53 M56 M56 M53 A5117 A41 A557 A49 A55(T) A561 A41(T) A54 A41 A557 A5116 A533 A56 A51 A41 A562 A494(T) A51 A55(T) A550(T) A494(T) A56(T) A41 A494(T) A5117(T) A483 A55(T) A5115 A557 A55(T) R i v e r D e e / A f

  • n

D y f r d w y Shropshire Union Canal S h r

  • p

s h i r e U n i

  • n

C a n a l Manchester Ship Canal R i v e r W e a v e r M a n c h e s t e r S h i p C a n a l W e a v e r N a v i g a t i

  • n

R i v e r A l y n / A f

  • n

A l u n B r i d g e w a t e r C a n a l

19 / 28

slide-21
SLIDE 21

Our urban regions

  • #
  • !
! ! ! ! ! ! !
  • RIVER

MERSEY RIVER DEE / AFON DYFRDWY Delamere Forest Ince Banks WIRRAL Gayton Sands White Sands Dungeon Banks Eastham Sands Dawpool Bank Bagillt Bank Sealand Sandycroft Mancot Royal Bretton Broughton Shotton Hawarden Drury Ewloe Northop Northop Hall Weston Hale Eastham Bromborough Raby Thornton Hough Barnston Gayton Flint Mountain Soughton/Sychdyn Hale Bank Grassendale Port Sunlight Storeton Pensby Irby Thurstaston Queensferry New Brighton Wepre Saltney Parkgate Ness Burton Puddington Shotwick Saughall Mollington Capenhurst Ledsham Willaston Hooton Childer Thornton Backford Stoak Wervin Picton Bridge Trafford Dunham-on-the-Hill Thornton-le-Moors Hapsford Alvanley Elton Ince Manley Mouldsworth Kingsley Newton Helsby Ashton Hayes Oscroft Duddon Burton Clotton Utkinton Willington Corner Kelsall Upton Mickle Trafford Great Barrow Tarvin Rowton Waverton Christleton Handbridge Littleton Guilden Sutton Blacon Lache Vicarscross Quarrybank Connah's Quay Buckley Speke Bebington Flint Garston Heswall Runcorn Neston Frodsham

176 158 14 10 8 7 4 15

A5032 A549 A56 A55(T) A5104 A540 A548 A5032 A540 A5119 A533 A551 A5480 A5137 A5117 A540 A548 A5117 A550 A41 A5119 A5032 B5463 B5439 B5132 B5125 B5132 B5130 B5125 B5132 B5132 B5153 B5151 B5137 B5134 B5132 B5129 B5155 B5393 B5137 B5128 B5136 B5132 B5171 B5125 B5132 B5152 B5151 B5128 B5133 B5135 B5126 B5136 B5138 B5463 B5151 B5155 B5441 M56 M56 M53 M53 M56 M53 M56 M56 M53 A5117 A41 A557 A49 A55(T) A561 A41(T) A54 A41 A557 A5116 A533 A56 A51 A41 A562 A494(T) A51 A55(T) A550(T) A494(T) A56(T) A41 A494(T) A5117(T) A483 A55(T) A5115 A557 A55(T) R i v e r D e e / A f

  • n

D y f r d w y Shropshire Union Canal S h r

  • p

s h i r e U n i

  • n

C a n a l Manchester Ship Canal R i v e r W e a v e r M a n c h e s t e r S h i p C a n a l W e a v e r N a v i g a t i

  • n

R i v e r A l y n / A f

  • n

A l u n B r i d g e w a t e r C a n a l

20 / 28

slide-22
SLIDE 22

How does it scale?

O(N) + O(3N) + O(N log N) + O(N log N) + O(N) +O(N) + O(N log N + s) ≈ O(N log N + s) (1) where N is the size of the original input and s is number of intersections. O(c · n2 log n) + O((n 2 + 2) · n log n) +O(n) + O(n2) ≈ O(c · n2 log n) (2) where n the much reduced size after model generalisation and c is the number of angular changes.

21 / 28

slide-23
SLIDE 23

Open problems

22 / 28

slide-24
SLIDE 24

Open problems

GI comparison with building simplification techniques GI administrative subdivisions and seed points GI acute exteriour angles CG inner rings, intersection free CG optimal solutions?

23 / 28

slide-25
SLIDE 25

Thank you for your attention! Thanks to Sheng Zhou, Nico Regnauld and Patrick Revell!

24 / 28

slide-26
SLIDE 26

Individual bits

d1 d

2

Pi-1 Pi+2 Pi+1 Pi-2 Pi Pnew

β α

25 / 28

slide-27
SLIDE 27

Individual bits

26 / 28

slide-28
SLIDE 28

Individual bits

27 / 28

slide-29
SLIDE 29

Individual bits

28 / 28