Y. Chenevoy [email protected] 1 © A. Belaïd [email protected] Constraint Propagation vs Syntactical Analysis for the Logical Structure of Library.

Y. Chenevoy [email protected] 1

© A. Belaïd [email protected]

Constraint Propagation vs Syntactical Analysis for the Logical Structure of Library References

A. BelaïdLORIA-CNRS Nancy France

Y. ChenevoyCRID Univ. Bourgogne Dijon, France

Outline• Structure Modeling

• Syntactical Analysis

• Constraint Propagation

• Results & Conclusion



Model: generic structure



Model: Attribute Grammar

Object ::= Constructor {subordinate objects [qualifier]}sequence, required,aggregate, optional,choice repetitive

Separator : space, graphic line / punctuation

Attributes : Physical Logical Typographical position lexicon typeface…

Weights : Attributes Sub-objects Imp / Reco. Imp / Hyp. Ambig.



• Top-down: Model driven

• Bottom-up: Data driven

• Mixed:

- Anchor points extraction (o)- Bottom-up: Choice of a rule

A o o ’o

- Top-down: verification for

left context o

right context ’o

- Add A to anchor points

Syntactical Analysis: the approach

’0

a1 … ai-1 ai … o … aj aj+1 … an

S

’A



Syntactical Analysis: Left context verification



Initials & Finals

Finals

O ::= Cho A B C F(O) = {A, B, C}O ::= Seq A B C F(O) = {C}O ::= Seq A B C? F(O) = {B, C}

O Vt , F*(O) = O

O Vn , F*(O) = F(O) (iF(O) F*(i))

Initials

O ::= Cho A B C I(O) = {A, B, C}O ::= Seq A B C I(O) = {A}O ::= Seq A? B C I(O) = {A, B}

O Vt , I*(O) = O

O Vn , I*(O) = I(O) (iI(O) I*(i))

Model : G = (Vn, Vt, P, S)



Indices Extraction: without OCR

Specific problems

4.7%

76.7%

37.5%

55.5%

37.5% 61%

31.5%

91.0%

43.3%

16.1%

Corr. with Corr. with



Indices Extraction: the approaches

Masks

Profile Projection

Bounding Box& Baseline

Sound Lines

- Projection- Spacing- Bounding Box

Bounding Box& Baseline

_-

. , ; :

Particular words

Text style (Bold Italic Underlined) ( spaced text) (Small text)

( ) {} []



Constraint Propagation



Neighbors (Example)



Propagation Results

Frag. Possible labels After Cons. Prop.

1 2 1

2 23 1

3 23 2

4 23 3

5 2 1

6 7 1

7 10 1

8 7 1

9 3 1

...

Anchor Points

Anchor Points



Model Compilation

• Pre-processing of the model

• Find initials, finals and neighbors

let LNa,p = the set of possible neighbors at the left of a in the rule :

p a (Vt Vn)* ((Vt Vn)* - {a})if a then LNa,p = F else LNa,p = F LNa

by extension ln*a,p = lLNa,p F*l

and LN*a = pPa ln*

a,p the left neighborhood of a in the model

A is left compatible with B if B LN*A or A RN*

B or(A B) PA PA and PB PB / PA PB



Results

Group Vedette:

Area Title:Principal Title:

End of the title:

Area Address / Date:

Address:Date:

Area Collection:

Group Cote:

Crossing Title:

Cros. Formulae:

Crossing Title:

200 references75%



Results: scientific references

400 references99.8%



Results

[Yua 95] J. Juan, Y. Y. Tang, and C. Y. Suen. Four Directional Adjacency Graphs (fdag) and their Application in Locating \34elds in Forms. In Third International Conference on Document Analysisand Recognition (ICDAR’95), pages 752\25 755. IEEE Computer Society Press, Aug. 1995.

Author(3) : J. Juan, Y. Y. Tang, and C. Y. SuenTitle : Four Directional Adjacency Graphs (fdag) and their Application in Locating fields in FormsEditor (0) :Month : AugYear : 1995Volume : Number : Publisher : IEEE Computer Society PressADDRESS : PA--GES : 752-755Organization: Booktitle : Third International Conference on Document Analysis and Recognition (ICDAR’95)Series :Note :



Conclusion

Weak points

• 25 % lead to inconsistant chain

• Feasability study without OCR

• Weakness of indices extractio algo.

• Local context handling

Strong points or improvements

• Fast analysis

• Structure well recognized for the others

• The method can be applied with OCR with better results

• Global context can be applied (path consistency) at the cost of CPU time

• Good for ambiguous models

• Limit the number of hypotheses during the analysis

• Limit the number of backtracking

Y. Chenevoy [email protected] 1 © A. Belaïd [email protected] Constraint Propagation vs Syntactical Analysis for the Logical Structure of Library.

Documents