|
Computer Vision 412 (233.412)
6 points / Semester 1
Handbook Description
The official Handbook entry Unit Aims
The aim of this course is to introduce you to image data, that is
information represented as an image, and to present methods for
processing and interpreting that data.
The course will show how to apply knowledge from areas such as
psychophysics, signal processing, projective geometry, linear algebra,
calculus, set theory, and information theory in order to develop
computational techniques for image understanding. Material presented
throught the course will be linked with current research in computer
vision.
You will learn how to program in Matlab and how to use Matlab's
Image Processing Toolbox.
Teaching Staff
| Lecturer |
Assoc. Prof. Ryszard Kozera |
Rm 1.1 |
| Email |
ryszard@csse.uwa.edu.au |
Recommended Reading
Gonzales/Woods, Digital Image Processing, Addison Wesley, 1992.
Vishvjit S. Nalwa, A Guided Tour of Computer Vision, Addison Wesley, 1993.
Ramesh Jain, Rangachar Kasturi and Brian Schunk, Machine Vision, McGraw Hill 1995.
Berthold Klaus Paul Horn, Robot Vision, MIT Press, 1986.
Olivier Faugeras, Three-Dimensional Computer Vision, MIT Press, 1993.
Contact Hours
With regards to formal contact hours, you will be required to attend
26 hours of lectures and a weekly 2 hour laboratory session.
To complete
the work in this unit you are expected to work out of hours. As this
is a 6 point unit, a total of 10 hours per week on average (including
lectures, research, preparation and revision) would be usual.
| Type |
Time |
Day |
Location |
| Lecture |
3-5 |
Wed |
General Purpose Building 2 Lecture Theatre |
| Lab |
9-11 |
Thur |
Computer Science Lab 2.05 |
| Lab |
11-1 |
Thur |
Computer Science Lab 2.05 |
Lecture Details
The lectures are presented in a 2 hour session (with a 15min break) each week.
The topics that will be covered in each week are listed below.
| Topic |
Details |
| 1 |
Introduction, image formation, camera models and perspective geometry. |
| 2 |
Binary images: thresholding, moments, topology. |
| 3 |
Connected components, and morphological operations. |
| 4 |
Fourier transform theory |
| 5 |
Image enhancement |
| 6 |
Edge detection: classical methods |
| 7 |
Camera calibration |
| 8 |
Projective geometry: 3D reconstruction from perspective cues |
| 9 |
Stereo and structured light |
| 10 |
Motion |
| 11 |
Colour |
| 12 |
Feature detection via phase congruency (or shape from shading) |
| 13 |
Interpreting line drawings (or shape from shading) |
Assessment
The assessment scheme for 233.412 consists of a submitted written
portfolio of practical work and a 2 hour examination paper. The
portfolio is in fact constructed as a web document, and is intended to
document the laboratory work done in the course. It will include:
brief descriptions of the work done,
images showing the output from various image processing operations
along with brief comments on the results, and will include the code,
or links to the code, written by the student. Code will be assessed
on the basis of its correctness, design and clarity, error trapping,
and documentation.
| Assessment |
% of final mark |
Assessment Dates |
| Portfolio |
30 |
Thursday week 13 |
| Exam |
70 |
June exam period |
Unsatisfactory Progress
Any student who does not demonstrate satisfactory progress in this
unit, as defined in the FECM
Policy on Assessment Practices and Procedures, may be refused admission to the
final examinations. The final deadline for notification of unsatisfactory progress is the
last day of Week 10.
Penalties
The School of Computer Science and Software Engineering has adopted a policy on
minimum penalties for late items of assessment.
This is the default policy of all units unless indicated otherwise, in writing, by the
specific unit coordinator.
This policy shall apply to all items of continuous assessment, whether
submitted either physically or electronically. Immediately after the submission deadline for an item of continuous
assessment, a penalty of 20 percent will be applied PER DAY or PART THEREOF.
The minimum mark possible for late submission is zero. The
percentage is based on the item´s total contribution to the unit´s
assessment. For example, a project contributing 40% to the unit´s
assessment will incur a penalty of 8 marks for each day late until it is submitted or
a mark of zero results.
A more detailed description is given in this School´s Policy on
Late Submission. The Faculty does have an appeals procedure, the details of which can found at the Policy for Appeals.
All work submitted for assessment must be the individual
student's own work. Any instance of plagiarism will be
brought to the attention of the Department Disciplinary Board
and dealt with under the Department's and Faculty's plagiarism
policies.
Plagiarism
Plagiarism is broadly defined to be when any portion
of the work presented for assessment, can be attributed
to another party. The student making the submission should acknowledge
what aspects of the presented work is not directly derived by
them. For the purposes of plagiarism it is irrelevant that you
have been given permission by someone to copy their work
and present it as your own.
You are directed to the School of Computer Science and Software Engineering Policy on Plagiarism and the Faculty of Engineering,Computing and Mathematics Policy on Plagiarism.
Faculty Scaling
Final assessment is subject to the Faculty Scaling Policy.
This information is correct as at 28-Feb-2005, but is subject
to change from time to time. In particular, The University
reserves the right to change the content and/or method of
presentation and/or the method of assessment of any unit of
study, to withdraw any unit of study or programme, and/or to
vary arrangements for any programme.
Copyright© 2005
School of Computer Science, & Software Engineering
The University of Western Australia
CRICOS Provider Code: 00126G
Last updated: 28-Feb-2005 |
