Neural object classification by pattern  recognition of one dimensional data arrays which represent object information tra...
Outline <ul><li>Definition and aim </li></ul><ul><li>State of the art in neural pattern recognition </li></ul><ul><li>Shap...
Definition and aim <ul><li>Pattern </li></ul><ul><ul><li>Groups of measurements or observations, defining points in an app...
Grasping Process 7-DOF manipulator performs with whole arm grasping of a planar object  <ul><li>Hyper-redundant manipulato...
Classification process Classification shape = [pi/4 pi/4 pi/4 pi/4 pi/4 pi/4 pi/4 pi/4  pi/4 pi/4 pi/4 pi/4 pi/4 pi/4] FEA...
<ul><li>Fourier Descriptors </li></ul>Turning Functions A special property of Fourier descriptors is that a shape’s symmet...
Neural Object Classification Artificial neuron Radial basis function Input Neuron 1 x   u F() 2 x u y    b w x u i i ...
<ul><li>A probabilistic neural network structure is able to classify the objects for tentacle case problem </li></ul><ul><...
Thank  you  for your attention.
Upcoming SlideShare
Loading in …5
×

Neural object classification by pattern recognition of one dimensional data arrays which represent object information transformed by nonlinear functions.

836 views

Published on

Pattern
Groups of measurements or observations, defining points in an appropriate multidimensional space.

Pattern recognition
Aims to classify data (patterns) based on either a priori knowledge or on statistical information extracted from the patterns.
My aim is to classify 2-dimensional objects with the help of manipulator and by the use of pattern recognition.

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
836
On SlideShare
0
From Embeds
0
Number of Embeds
17
Actions
Shares
0
Downloads
13
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Neural object classification by pattern recognition of one dimensional data arrays which represent object information transformed by nonlinear functions.

  1. 1. Neural object classification by pattern recognition of one dimensional data arrays which represent object information transformed by nonlinear functions. presented by Kayhan Ince Thesis Supervisor: Univ.Prof. Dipl.-Ing. Dr.techn. FAVRE-BULLE, Bernard Thesis Co-Advisor: Dipl.-Ing. Fauaz Labadi ACIN – Automation and Control Institute
  2. 2. Outline <ul><li>Definition and aim </li></ul><ul><li>State of the art in neural pattern recognition </li></ul><ul><li>Shape analysis (classification) and grasping process </li></ul><ul><li>Simulation Results </li></ul><ul><li>Conclusion </li></ul>
  3. 3. Definition and aim <ul><li>Pattern </li></ul><ul><ul><li>Groups of measurements or observations, defining points in an appropriate multidimensional space. </li></ul></ul><ul><li>Pattern recognition </li></ul><ul><ul><li>Aims to classify data (patterns) based on either a priori knowledge or on statistical information extracted from the patterns. </li></ul></ul><ul><ul><li>My aim is to classify 2-dimensional objects with the help of manipulator and by the use of pattern recognition. </li></ul></ul>
  4. 4. Grasping Process 7-DOF manipulator performs with whole arm grasping of a planar object <ul><li>Hyper-redundant manipulator </li></ul><ul><ul><li>- Serial-Chain-Mechanism </li></ul></ul><ul><ul><li>- Planar, rotational Joints </li></ul></ul><ul><ul><li>n Degrees of Freedom (DOF) </li></ul></ul><ul><ul><li>Array of angles </li></ul></ul>joints Base Link
  5. 5. Classification process Classification shape = [pi/4 pi/4 pi/4 pi/4 pi/4 pi/4 pi/4 pi/4 pi/4 pi/4 pi/4 pi/4 pi/4 pi/4] FEATURE VECTOR DRAWTESTSET TURNINGFUNCTION PERIODIC CLASSIFICATION RESULTS CLASSIFY
  6. 6. <ul><li>Fourier Descriptors </li></ul>Turning Functions A special property of Fourier descriptors is that a shape’s symmetry shows up in the feature vector. <ul><ul><li>Arkin published an efficient method for comparing polygonal shapes. The notion of the turning function which represents the shape of an object. </li></ul></ul>Classification PNN
  7. 7. Neural Object Classification Artificial neuron Radial basis function Input Neuron 1 x   u F() 2 x u y    b w x u i i 1 w 2 w b Output(classification)
  8. 8. <ul><li>A probabilistic neural network structure is able to classify the objects for tentacle case problem </li></ul><ul><li>Turning functions carry the distance information of the objects </li></ul>Conclusion
  9. 9. Thank you for your attention.

×