TITLE: ProcessRDR and Beyond

PRESENTER: Avishkar Misra, CSE PhD Student

AFFILIATION: School of Computer Science & Engineering and National ICT Australia

DATE: Monday 25th October 2004

TIME: 11am - 12noon

PLACE: CSE K17 1st Floor Seminar Room

ABSTRACT:

Computer vision systems have a strong dependence on the Domain Knowledge in order to work effectively. This idea has been widely accepted and spurred research in combing learning and knowledge engineering methods to vision systems. The majority of this effort has largely neglected to address the problems associated with learning Control Knowledge. Control knowledge pertains to learning "when and how" a particular image processing procedure is applied.

We have developed a framework called ProcessRDR, which we feel address some of these problems. ProcessRDR applies the concepts of Ripple Down Rules, to complex vision systems. The technique has been applied to solve a medical imaging problem of Lung Boundary extraction from High Resolution Computed Tomography (HRCT) of the lung. The results have been promising and opened new interesting directions.

This talk will provide an overview of ProcessRDR as well as the potential directions of this work. I will also discuss the challenges that need to be addressed and the existing strategies that may help.

BIOGRAPHY OF SPEAKER:

Avishkar Misra is a 2nd year PhD student at CSE. He holds a Bachelors of Engineering in Software from CSE. For more information please consult his website:

www.cse.unsw.edu.au/~amisra

HOST

Associate Professor Sowmya Arcot
School of Computer Science and Engineering
University of New South Wales
somya"at"cse.unsw.edu.au
+61 2 9385 6933 (Internal: x56933)

SEMINART CONVENER:

Bernhard Hengst
National ICT Australia, Symbolic Machine Learning & Knowledge Acquisition
Tel: +61-2-9385-6940