Fourier descriptors - PowerPoint PPT Presentation

About This Presentation
Title:

Fourier descriptors

Description:

Fourier descriptors Ahmad Alkhodre & Mostefa Belarbi Insa Lyon Georg Langs & Srdan Tosovic Vienna University Nicolas Sicard Pavia University – PowerPoint PPT presentation

Number of Views:62
Avg rating:3.0/5.0
Slides: 19
Provided by: SimonA85
Category:

less

Transcript and Presenter's Notes

Title: Fourier descriptors


1
Fourier descriptors
  • Ahmad Alkhodre Mostefa Belarbi
  • Insa Lyon
  • Georg Langs Srdan Tosovic
  • Vienna University
  • Nicolas Sicard
  • Pavia University
  • Lecturer Alberto Aguado
  • University of Surrey

2
Contents
  • Introduction
  • Fourier descriptors
  • Classification
  • Occlusion
  • Suggestions
  • Conclusion

3
Introduction
  • Using shapes
  • Advantages
  • Drawbacks

4
Fourier decomposition
?

...
5
Fourier descriptors
decomposed to real imaginary part
original picture
border
Fourier coefficients
Fourier descriptors
6
Fourier decomposition(of Nicolas)
Processing
Processing
7
Invariance (Rotation)work in progress
8
Reconstruction (of Nicolas)
9
Classification
  • Database of Fourier descriptors with 14 airplanes
    and 1 Nicolas.
  • Evaluation function for matching the test object
    with one object or Nicolas in the database.
  • Astonishing results

10
Occlusion problem
  • Random results, when classifying occluded curves
  • How much is occluded?
  • Which part of the original curve?
  • Which airplane?

11
Classification occlusions
database building
Create databases of subcurves for each airplane
Alpha-beta decomposition
12
Curve represented by subcurve descriptors
database building
1.
2.
3.
4.
beta
alpha
5.
13
Processing of input
matching
Process the curve as an interval from a to b, for
every (a,b) couple present in the database
n x
Store results in a datastructure similar to the
database
14
Matching
matching
Input image
15
Search for lowest point
matching
a
b
16
Results
matching
inputshape
match
b
a
17
Problems, observations
matching
  • Symmetry
  • Descriptors are invariant against mirroring
  • Time for more testing
  • Tests were performed on one airplane
  • Matching is dependent on number of airplanes in
    the database
  • Weigthing algorithm for lowest point search
  • Robustness

18
Conclusion
  • What we have done
  • Shape extraction
  • Fourier coefficients
  • Invariant Fourier descriptors
  • Shape reconstruction
  • Classification
  • Proposition for occlusion problem
  • What we still need to do
  • Generalization of alpha-beta decomposition
Write a Comment
User Comments (0)
About PowerShow.com