Canonical forms and orbit identification problems in machine vision

S. Takahashi, B. K. Ghosh

Research output: Contribution to journalConference articlepeer-review


We introduce canonical forms for perspective dynamical systems under the action of a perspective group and illustrate their application to parameter identification with the aid of a single CCD camera. We show that the parameters in the canonical form can be identified uniquely using an Extended Kalman Filter (EKF).

Original languageEnglish
Pages (from-to)5175-5181
Number of pages7
JournalProceedings of the IEEE Conference on Decision and Control
StatePublished - 2000
Event39th IEEE Confernce on Decision and Control - Sydney, NSW, Australia
Duration: Dec 12 2000Dec 15 2000


  • Canonical forms
  • Extended Kalman filter
  • Identification
  • Kronecker index
  • Machine vision
  • Parameter
  • Perspective dynamical systems


Dive into the research topics of 'Canonical forms and orbit identification problems in machine vision'. Together they form a unique fingerprint.

Cite this