From Web Documents to Old Books, Works in Progress in Graphics Recognition

1
From Web Documents to Old BooksWorks in Progress
in Graphics Recognition

• Mathieu Delalandre
• Meeting of Document Analysis Group
• Computer Vision Center
• Barcelona, Spain
• Thursday 23th November 2006

2
Plan

• Short CV
• Vector Graphics Indexing and Retrieval
• Dropcap Image Retrieval

3
Short CV

• Personal Information
• Mathieu Delalandre, 32 years old
• Academic Degrees
• 1995-1998 Lic.Sc in Electronic
• Rouen University, France
• 1998-2001 M.Sc in Industrial Computing
• Rouen University, France
• Research Periods
• Length Position Laboratory Subject
• 6 months Master LITIS symbol recognition
• 3 ½ years PhD LITIS drawing understanding
• 5 months Post-doc SCSIT vector graphics
  indexing
• months Post-doc L3i dropcap image retrieval
• 2 months Contract LITIS performance
  evaluation
• 3 years Post-doc CVC

4
Plan

• Short CV
• Vector Graphics Indexing and Retrieval
• Dropcap Image Retrieval

5
Vector Graphics Indexing and Retrieval
Application of vector graphics 1982 Computer
Aided Design (DXF 1982) 1985 Office software
(PS 1985, CGM 1987, WMF 1993) 1996 Web (PNG
1996, SVG 2001 ..)

• Vector graphics are growing on Web
• SVG 1.0
• SVG widely used
  structured documents Mong03,
  geographic maps Chen04, technical drawings
  Kang04
• 2005 Powerful editors (Inskape, Webdraw, )
• Internet Explorer and Mozilla Firefox support SVG

6
Vector Graphics Indexing and Retrieval

• System overview
• Doer98 Tom03
• Look like pattern recognition approach

7
Vector Graphics Indexing and Retrieval
You see 5
You have 9
We need a clean-up
8
Vector Graphics Indexing and Retrieval

• Our approach (next)

Time processing on Mikado database
9
Vector Graphics Indexing and Retrieval
To work on retrieval engine now ? How to evaluate
the retrieval results after ? We must work on
performance evaluation before ?
How to get the ground truth ? Produce ground
truth from existing document take time, we must
produce synthetic document.
10
Vector Graphics Indexing and Retrieval
Graphical Objects
Low Level Primitives

• General rules
• object number
• document size
• object choice
• -probability distribution
• -rotation and scale range
• -position constraints
• -overlapped or not

• Domain rules
• must be connected
• must be adjacent
• must be include
• can include

• Noise rules
• to scale line
• to broke line
• to move line

(4) To move objects according to domain rules
(5) To delete oldest
alone objects cycle number
II
Vector Graphics
(1) To insert a new object while underhand object
number (2) To move other
objects if it cant do (1) (3) To exit if it
cant do (1) and (2), then run (4) and (5)
while
I
III
(6) Adding noise on low level primitives
composing objects
Ground Truth
In progress
11
Vector Graphics Indexing and Retrieval

• Works done
• Fast graph building from vector graphics
• Production of first synthetic documents
• Works in progress
• To produce more complex synthetic documents
• To work on model selection
• To work on index structuration

About project dot-line 04/05 SCSIT Post
doc 02/06 IRCSET Application A. Winstanley
(NCG, Dublin University) 04/06 Eureka Meeting
eConnector, HP Lab 06/06 ANVAR
Application informal agreement 11/06 EPEIRES
contract 2007 To visit A. Winstanley (NCG,
Dublin University) To take contact with M.
Fonseca (IST, Lisbon University) 2008 JM Ogier
plan to mount a European project
12
Plan

• Short CV
• Vector Graphics Indexing and Retrieval
• Dropcap Image Retrieval

13
Dropcap Image Retrieval

• Old books of XV and XVI centuries

Which part and kind of graphics in old books
Book 46
Page 1385
Graphics 4755 (3.4 per page)
Foreground pixel Jour05 63 textual 37 graphical
Graphics type 41 dropcap 59 others
CESR Database
Old Graphics
14
Dropcap Image Retrieval
In what are interested historian people with
these images ?
Why ?
15
Dropcap Image Retrieval
Which descriptor use ?
16
Dropcap Image Retrieval
17
Dropcap Image Retrieval
Digitalization problems Lawrence00 Several
image providers Several digitalization tools Long
process Human supervised Complex post-processing
plate-form
Contrôle
18
Dropcap Image Retrieval
Compression results
19
Dropcap Image Retrieval
We can do it in an easy way by comparing
foreground histogram

• Centering

20
Dropcap Image Retrieval
21
Dropcap Image Retrieval
Selection results
22
Dropcap Image Retrieval
23
Dropcap Image Retrieval

• Works done
• QUEID to filter and analyse image database
• Speedup comparison using two feature
• RLE compression
• System approach
• Works in progress
• To add operator to improve system
• To extend our system to produce benchmark
  database

About project dot-line 09/05 MADONNE
Postdoc 06/06 1er CESR Technical
Meeting 09/06 ANAGRAM Worshop (Fribourg) 10/06 2
sd CESR Technical Meeting 10/06 NaviDoMass
agreement 2007 GDR-JC Project (LMA, LI, CreSTIC,
LITIS, CVC) To put online the system on CESR
website old graphic working group (Glasgow,
Tours )
24
Bibliography

1. J. Mong and D. Brailsford. Using svg as the
   rendering model for structured and graphically
   complex web material. In Symposium on Document
   Engineering (DocEng), pages 88-91, 2003.
2. Y. Chen, J. Gong, W. Jia, and Q. Zhang. Xml-based
   spatial data interoperability on the internet. In
   Conference of International Society for
   Photogrammetry and Remote Sensing and Spatial
   Information Sciences (ISPRS), pages 167-201,
   2004.
3. J. Kang, B. Lho, J. Kim, and Y. Kim. Xml-based
   vector graphics Application for web-based design
   automation. In International Conference on
   Computing in Civil and Building Engineering
   (ICCCBE), pages 170-178, 2004.
4. M. Weindorf. Structure based interpretation of
   unstructured vector maps. In Workshop on Graphics
   Recognition (GREC), volume 2390 of Lecture Notes
   in Computer Science (LNCS), pages 190-199, 2002.
5. N. Journet, R. Mullot, J. Ramel, and V. Eglin.
   Ancient printed documents indexation a new
   approach. In International Conference on Advances
   in Pattern Recognition (ICAPR), volume 3686 of
   Lectures Notes in Computer Science (LNCS), pages
   513-522, 2005.
6. V. D. Gesu and V. Starovoitov. Distance based
   function for image comparison. Pattern
   Recognition Letters (PRL), 20(2)207-214, 1999.
7. S. Loncaric. A survey of shape analysis
   techniques. Pattern Recognition (PR),
   31(8)983-1001, 1998.
8. G. Lawrence and al. Risk management of digital
   information A file format investigation. RLG
   DigiNews, 8(4), 2000.