Key words
Image processing, Computer vision, Stereo vision
Objectives
The contents of this course is complementary to the contents of the courses on Multimedia and Computer Graphics. Since the course does not use concepts of these two courses and the course on signal theory, it can be folowed by students that did not follow one of these three courses. The course looks at modern, advanced techniques for image processsing and computer vision such as face recognition, understanding of outdoor scenes and conversion of 2D images into 3D, the use of intelligent camera' for surveillance. The emphasis is on algorithms and the programming skills that are necessary for the implementation of complex algorithms in computer vision. The environment used is Matlab.
Topics
1. Introduction: overview of applications in 2D and 3D computer vision and introduction to pattern recognition
2. Datareduction and techniques for data fitting
3. Epipolar geometry and camera parameters
4. Morphological image processing: dilation, erosion, opening, closing, algorithms, extension to grey values
5. Image segmentation: edge detection, line detection, Hough transform, RANSAC-algorithm, region segmentation, watershed-algorithm.
6. Complexity of image processing algorithms
7. Graph algorithms and computer vision
8. Search algorithms
9. Representation and description of images: chain codes, skeletons, shape, texture.
10 Object recognition: feature vectors, decision theory, neurale networks, structural models
Prerequisites
Good basis of programming and linear algebra.
Final Objectives
Materials used
::Click here for additional information::
Manual: "Digital Image Processing," Gonzalez and Woods, Prentice Hall, 2002.
Background material epipolar geometry:
"Multiple View geometry in computer vision," Hartley and Zisserman, Cambridge University Press, 2003.
Study costs
60 euro for book
Study guidance
Teaching Methods
Assessment
Lecturer(s)
|
|
