Introducing TRIGRAPH trimodal writer identification

1
Introducing TRIGRAPHtrimodal writer
identification
Ralph Niels, Louis Vuurpijland Lambert
Schomaker?
Dutch Forensic Institute
? Artificial Intelligence Institute University
of Groningen
Nijmegen Institute for Cognition and
Information Radboud University Nijmegen
ENFHEX conference - November 2005 Budapest,
Hungary
2
Overview

• Computer assisted document examination
• TRIGRAPH combines 3 methodsI Automatic
  features from imageII Manually measured
  propertiesIII Allographic features
• Recent achievement intuitive matching
• Summary
• Next steps

3
Computer assisted document examination
4
Computer assisted document examination
5
Improving on current systems

• Systems do not benefit from recent advances in
  pattern recognition and image processing
• New insights in
• automatically derivedhandwriting features
• user interface development
• innovations in forensic writer identification
  systems
• Aim Suspected document in top-100 hit list from
  database of gt 20,000 writers

6
Design requirements

• Improve on currently available performance
• Minimize amount of manual labor
• Exploit human cognition and expertise
• Correspond to expectations of human experts

7
WANDA

• Integrate techniques in WANDA Workbench(Franke
  et al., ENFHEX News 2004 Van Erp et al., JFDE
  (16) 2004)

8
Three approaches

• I Automatic features from images
• II Manually measured properties
• III Allographic features

9
Automatic features from images (1)
I

• Layout and spacing
• Ink morphology
• (Franke)

10
Automatic features from images (2)
I

• Local shape (Bulacu)

11
Automatic features from images (3)
I

• Grapheme-fraglet tables (Schomaker)

12
Manually measured properties
II

• Fish
• Script
• Wanda

13
Allographic properties (1)
III

• (Vuurpijl, Niels) Matching characters by
• Considering global shape characteristics
• Reconstructing and comparing production process
• Zooming in on particular features

14
Intuitive matching (1)
III

• Given 2 dynamic trajectories(one questioned,
  one from aset of prototypes)
• Technique Dynamic TimeWarping
  (point-to-pointcomparison)
• Result similarity measure thatcan be used to
  find prototypethat is most similar toquestioned
  sample

15
Intuitive matching (2)
III

• Experiment compare various techniques
• Result Dynamic Time Warping yields visually
  convincing (or intuitive) results
• Our work on DTW was previously presented at

• 9th International Workshop on Frontiers in
  Handwriting Recognition(IWFHR-2004), Japan.
• 12th Conference of the International
  Graphonomics Society(IGS-2005), Italy.
• 8th International Conference on Document
  Analysis and Recognition(ICDAR-2005),
  South-Korea.

16
Allographic properties (2)
III

• (Semi-)automatic extraction of dynamic
  information
• Automatically extract traces from scanned
  document
• Verify resulting trajectories with allograph
  prototypes
• Start user-interaction in case of confusion
• Advantages
• More reliable measurements
• Online character recognition techniques
• Search for particular allographs in documents
• Visually convincing matching techniques

17
Summary

• Computers can help forensic experts in measuring
  handwriting and searching databases
• In TRIGRAPH, new insights from different
  scientific areas will be used
• In TRIGRAPH, new UI methods will be combined with
  techniques developed in three modalities I
  Automatic features from images II Manually
  measured properties III Allographic features

18
Next steps

• Automatic extraction of dynamical information
  from scanned images
• Supervised character segmentation
• Allograph based verification of results

19
Introducing TRIGRAPHtrimodal writer
identification
Ralph Niels, Louis Vuurpijland Lambert
Schomaker?
? Artificial Intelligence Institute University
of Groningen
Dutch Forensic Institute
Nijmegen Institute for Cognition and
Information Radboud University Nijmegen
Questions?
ENFHEX conference - November 2005 Budapest,
Hungary