Document Analysis Systems

1
Document Analysis Systems
Bedola Roberto, Bordoni Davide, Franc
Vojtech Supervised by Luca Lombardi

• Overview
• Introduction
• Our implementation of Wahl et al framework
• Multiresolution Approach for page segmentation
• Conclusions

2
www.ip2001.5u.com

• Id ip2001.5u.com
• Pass 2001ip

3
Introduction

• Document understanding
• Extraction of relevant information from documents
  (letters, forms, engineering drawings, etc. )

Graphics Processing
Digitalized Image
Page Segmentation
Textual Processing (OCR)
4
Wahl Our implementation
Segmentation
Connected Component Extraction
Region merging
Block Characterization
Text extraction
5
Example 1
After binarization
After segmentation
Original doc.
6
Example 2
Labelled image Before merging
Labelled image After merging
Classified image
7
Multiresolution

• Principles
• The use of various filter permit to find the
  different zones.
• Techniques
• The use of some parameter permit to cut some
  pixel.

8
Multiresolution
9
Mean Variance

• Mean
• Variance

10
Background Condition

• We can find the background
• 240 lt Mean lt 255
• 225 lt Mean lt 255 image with noise
• 0 lt Variance lt 15
• 0 lt Variance lt 25 image with noise

11
Median Threshold

• Median
• Threshold

12
Image and Graphics

• 4 steps
• Middletone Segmentation
• Median filter
• Threshold filter
• Pixel counting
• If threshold ? 170 ? counter
• Classification
• If counter/area ? 0.7 ? Graphic
• else ? Pictures

13
Text

• 3 steps
• Segmentation of mean image
• Mean filter
• Pixel text counting
• If 240? mean ? 210 variance ? 10 ? counter
• Classification
• If counter/area ? 0.7 ? Text
• else if counter/area ? 0.3 ? analyse a more
  detailed image
• else ? Unclassifiable

14
Experimental Results
15
Conclusions

• Wahl et al. method
• Simple and accurate, but slower.
• We finish to implement this method
• Multiresolution
• Complex and faster, but need many parameters
• Only partial implementation

16
Goodbay
Meat or fish