After attending this course, the student will know and will able to:
- Apply fundamental signal processing techniques on 2D images using Matlab
- Identify and apply image restoration methods to filter image artefacts such as image noise, distortion and blurring
- Recognize and apply advanced image processing techniques including geometrical transforms, edge and line detection and morphological operations
- Describe image formation models such as geometric camera model and perspective projection
- Implementation of computer vision techniques such as camera calibration and rectification, and structure image analysis
- Describe and apply computer vision and image processing algorithms to solve a practical problem in Matlab and evaluate the performance of the proposed implementation
The course familiarizes students with digital image processing and computer vision techniques. It provides the fundaments for 2-D signal processing applied to digital images. It also discusses techniques for the extraction of 2D, 3D, or 4D information that is represented by a digital image (or image sequence). Examples of computer vision tasks are:|
a) the detection, e.g. checking the presence of an object or event.
b) The recognition or identification of an object or person.
c) The measurement of the parameters of an object, e.g. position, size, shape.
d) Motion analysis of objects.
The topics of the course include image formation and acquisition, 2D Fourier transforms, image operations, image segmentation, regional description, recognition and parameter estimation. The course involves practical work in which the students design a vision system for a simple application. As such, the student acquires programming skills using Matlab and its image processing toolbox. Examples of design tasks that students can select are:
a) Virtual advertising: inserting virtual advertising images into recorded movies of sports events
b) Motion analysis: tracking an object in a cluttered movie.
c) 3D face reconstruction from 3 images
d) 3D tracking of facial point features.
This course is mandatory for the follow-up course: 'advanced computer vision and pattern recognition'.
Content Image formation, image operations and image analysis.
Individual assignments (30%)
Group project + Oral discussion (50%)
Written exam (20%)
Assumed previous knowledge
|Master Electrical Engineering|
|Master Interaction Technology|
|Master Mechanical Engineering|
|Master Biomedical Engineering||Required materials|
|Lecture notes made available via website|
|Syllabus 3D Computer Vision (made available via website)|
|M. Sonka, V. Hlavac, R. Boyle: Image Processing, Analysis, and Machine Vision, 3rd edition, Thomson, 2008. This book will be used in the follow-up course 'advanced computer vision and pattern recognition'.|
|R.C. Gonzales, R.E. Woods: Digital Image Processing, 2nd edition, Prentice Hall, 2002.|
|D.A. Forsyth, J. Ponce: Computer Vision - a Modern Approach, Prentice Hall, 2003|
|Self study without assistance|
|Assignments, Group Project, Exam|