image processing and interpretation group university of nottingham eureka meeting, l3i laboratory,...

15
Image Processing and Interpretation Group University of Nottingham Eureka Meeting, L3i Laboratory, La Rochelle University Tuesday 20th April 2006 Fast Building of Region Graph from SVG Mathieu Delalandre, Zouba Karim, Norolala Ramangaseheno Supervisors Tony Pridmore (IPI, Nottingham University, UK) Eric Trupin (PSI, Rouen University, France)

Upload: richard-long

Post on 16-Dec-2015

214 views

Category:

Documents


1 download

TRANSCRIPT

Image Processing and Interpretation GroupUniversity of Nottingham

Eureka Meeting, L3i Laboratory, La Rochelle UniversityTuesday 20th April 2006

Fast Building of Region Graph from SVG

Mathieu Delalandre, Zouba Karim, Norolala Ramangaseheno

Supervisors Tony Pridmore (IPI, Nottingham University, UK)

Eric Trupin (PSI, Rouen University, France)

Image Processing and Interpretation GroupUniversity of Nottingham

Eureka Meeting, L3i Laboratory, La Rochelle UniversityTuesday 20th April 2006

<rect x="400" y="100" width="400“ height="200"fill="yellow" stroke="navy" stroke-width="10" />

(a) (b)

Common formats :

• AI (Adobe Illustrator)

• SVG (Scalable Vector Graphic)

• WMF (Windows Metafile)

• EPS (Encapsulted PostScript)

• DXF (AutoCAD)

•ClipArt

•Flash

Example

WMF pen EPS plane

IntroductionWhat are Vector Graphics ?

ClipArt Cheese

Image Processing and Interpretation GroupUniversity of Nottingham

Eureka Meeting, L3i Laboratory, La Rochelle UniversityTuesday 20th April 2006

IntroductionVector Graphics Indexing & Retrieval (1/2)

• Vector graphics are growing on Web and databases • [Mong’03] [Chen’04] [Kang’04] …

• Few I&R systems have been developed (> 2000)• [Love-01] [Sciascio’04] [Dosch’04] [Fonseca’05]

[Rusiñol’05] [Zakaria’05] …

Vector Graphics

Features Extraction Matching

Index Ranking

Image Processing and Interpretation GroupUniversity of Nottingham

Eureka Meeting, L3i Laboratory, La Rochelle UniversityTuesday 20th April 2006

IntroductionVector Graphics Indexing & Retrieval (2/2)

• Line Graph [Dosch’04] [Zakaria’05] …• Symbol recognition

• Region Graph [Fonseca’05] [Rusiñol’05] ..• Document indexing

• Our works region graph extraction for SVG Indexing and Retrieval large sized data a fast approach • Two steps : unformat SVG and region graph building

Image Processing and Interpretation GroupUniversity of Nottingham

Eureka Meeting, L3i Laboratory, La Rochelle UniversityTuesday 20th April 2006

Unformat SVGUnformat process, what is it ?

Example: overlapped rectangles

L1 L2

R3

R1

R2

L3L4

<rect x="400" y="100" width="400" height="200"fill="blue" /><rect x="650" y="200" width="400" height="200"fill="yellow" />

R2R1 R3

?

to unformat (or to broke) SVG

Image Processing and Interpretation GroupUniversity of Nottingham

Eureka Meeting, L3i Laboratory, La Rochelle UniversityTuesday 20th April 2006

filtering

Parsing for line extraction

SVG document

Unformat SVG Overview of our approach

set of (joined) lines set of no joined lines

Image Processing and Interpretation GroupUniversity of Nottingham

Eureka Meeting, L3i Laboratory, La Rochelle UniversityTuesday 20th April 2006

Unformat SVGWhy using a filtering step

you see 5 lines you have 9 lines

Image Processing and Interpretation GroupUniversity of Nottingham

Eureka Meeting, L3i Laboratory, La Rochelle UniversityTuesday 20th April 2006

Unformat SVG Our filtering process

l1 includes l2

l1 same as l2

l1 joins l2

(a)

(b)

(a)

(b)

(a)

(b)

l1 intersects and overlaps l2

l1 intersects only l2

Image Processing and Interpretation GroupUniversity of Nottingham

Eureka Meeting, L3i Laboratory, La Rochelle UniversityTuesday 20th April 2006

Unformat SVGExamples of results

crossing point merged lines

Image Processing and Interpretation GroupUniversity of Nottingham

Eureka Meeting, L3i Laboratory, La Rochelle UniversityTuesday 20th April 2006

Region Graph BuildingHow it works ?

K Lines Line Graph

Finding regions and their links

Image Processing and Interpretation GroupUniversity of Nottingham

Eureka Meeting, L3i Laboratory, La Rochelle UniversityTuesday 20th April 2006

Region Graph BuildingApproaches used in the literature

• Approaches based on graph handling [Fonseca’05] ..• region detection = problem of finding minimum length

cycles inside a graph

Image Processing and Interpretation GroupUniversity of Nottingham

Eureka Meeting, L3i Laboratory, La Rochelle UniversityTuesday 20th April 2006

• Based on [Weindorf’01] works : using vectorial information

Definition: direct angle = anticlockwise

Region Graph BuildingOur approach (1/3)

L1

L5L2L4

L3

L1

L4 L5 L2

L3

Vector Graphic document Line Graph

e

b

e

be

e

e: end of a Line

b: beginning of a Line α:Direct angle between 2 Lines

beeb

b

b

eb

b

ee

L1

L4 L2L5

L3

Specialized Line Graph

α(2-1)

α(1-2)

b2,e1e1,b2

e4,b1

b1,e4

Image Processing and Interpretation GroupUniversity of Nottingham

Eureka Meeting, L3i Laboratory, La Rochelle UniversityTuesday 20th April 2006

[Clementini’93]

a b a bab

b

a

adjacency overlap strict and tangential inclusion

Region Graph Building Our approach (2/3)

a b

neighboring

R1

Ymax

Ymax

Ymin

Ymin

Xmin Xmin Xmax Xmax

R2

[Xmin < Xmin & Xmax > Xmax] & [Ymin < Ymin & Ymax > Ymax]

Image Processing and Interpretation GroupUniversity of Nottingham

Eureka Meeting, L3i Laboratory, La Rochelle UniversityTuesday 20th April 2006

Region Graph Building Our approach (3/3)

Image Processing and Interpretation GroupUniversity of Nottingham

Eureka Meeting, L3i Laboratory, La Rochelle UniversityTuesday 20th April 2006

Conclusion and Perspectives

• Conclusion• First system dealing with unformating problems• First system allowing to build region graph from large sized

documents and from large sized databases

• Perspectives• Extend to curves and arcs processing • Extend built graphs with neighboring relations• Reduce the unformat complexity step using a zone sorting

algorithm• Use it for retrieval and indexing (not only graph building)