1. Fundamentals of Digital Imaging

Literature: GW1 pp 1 - 73, GW2 pp 1 - 63
Date: 22th of March
Presenter: Silver Leinberg
Slides: PDF
Demo materials: -

  • History
  • Various application areas
  • Human perception of images
  • Image sampling and quantization
  • Three classes of image processing operators

2. Reconstruction of 3d shapes form 2D images

Literature: Additional materials
Date: 1st of March
Presenters: Hans Mäesalu & Egon Elbre & Konstantin Tretjakov
Slides: PDF
Demo materials: ZIP

  • Reconstruction of 3D images form 2D image pairs

3. Intensity transforms and spatial filtering

Literature: GW1 pp 105-193, GW2 pp 34- 117
Date: 8th of March
Presenters: Hannes Tamme
Slides: PDF
Demo materials: ZIP

  • Fundamentals of spatial filtering: correlation and convolution
  • Vector and matrix representation of spatial filtering
  • Smoothing spatial filters: linear and order based
  • Sharpening spatial filters: Laplacian and gradeint

4. Fundamentals of colored imaging

Literature: GW1 pp 282-348, GW2 pp 194-241, P 39-88
Date: 15th of March
Presenters: Kristjan Krips & Timo Petmanson
Slides: PDF
Demo materials: ZIP

  • Physical model of colors
  • RGB, HSI, HCL color models
  • Transformations between color models
  • Color slicing and color correction

5. Frequency domain filtering

Literature: GW1 pp 199 - 310 GW2 pp 109 - 139
Date: 22th of March
Slides: PDF
Demo materials: ZIP (unpolished MATLAB code)

  • 1D and 2D Fourier transformation
  • Nyquist theorem
  • Discretization artifacts: aliacing and moire effects
  • Inverse Fourier transform
  • Low- and high-pass filtering
  • Bandreject and bandpass filters

6. Morphological processing of monochrome images

Literature: GW1 pp 627 - 679, GW2 pp 335 - 377, P pp 421 - 461
Date: 29th of March
Slides: PDF
Demo materials: ZIP

  • Erosion and dilation
  • Opening and closing
  • Boundary detection, hole filling and convex hulls
  • Extraction of connected components
  • Thinning, thickening and skeletons
  • Noise reduction methods

7. Line, edge, blob and corner detection

Literature: GW1 pp 689 - 738, GW2 pp 489 - 511, P pp 465 - 531
Date: 5th of April
Slides: PDF
Demo materials: ZIP

  • Edges and lines
  • Blobs and specs. Blob hierarchy
  • First and second order edge detection methods
  • Canny edge detection algorithm
  • Corner detection. Harris corner detection. Susan corner detection

8. Image segmentation

Literature: GW1 pp 738 - 787, GW2 pp 511 - 550, P pp 579 - 615
Date: 12th of April
Slides: PDF
Demo materials: ZIP

  • Thresholding. Optimal thresholding. Adaptive thresholding.
  • Thresholding based on image features.
  • Region-based thresholding. Region growing. Region splitting and merging.
  • Watershed algorithm for image segmentation.
  • Motion segmentation

9. Representation of detected objects

Literature: GW1 pp 795 - 857, GW2 pp 426 - 483, P pp 535 - 576
Date: 19th of April
Slides: PDF
Demo materials: ZIP

  • Boundary and its features
  • Object skeletons and its features
  • Topological features
  • Texture and its features

10. Elements of machine learning

Literature: separate materials
Date: 26th of April
Slides: PDF
Demo materials: ZIP

  • What is machine learning?
  • What is classification?
  • What are test and training errors?
  • What is holdout and crossvalidation methods
  • How to use Bayes classifier? How to use SVM?

11. Object recognition

Literature: GW1 pp 861 - 907, GW2 pp 484 - 511, P pp 651 - 677
Date: 3rd of May
Slides: PDF
Demo materials: Haar training tutorial (by courtesy of Hannes Tamme)

  • PCA. LDA
  • Feature extraction through correlation element
  • Haar training. Corresponding features. Cascades.

12. Tracking Moving Objects form Image Sequences

Literature: Josef Bigün: Vision with Direction: A Systematic Introduction to Image Processing and Computer Vision (2006)
Literature: Bernd Jähne, Horst Haussecker, Peter Geissler: Handbook of Computer Vision and Applications, Volume 3: Systems and Applications (1999)
Date: 10rd of May
Slides: PDF
Demo materials: ZIP

  • Errors and optical effects caused by motion
  • State updates with Markov and Kalman filters
  • Frequency domain signatures of moving objects
  • Model-based recognition of moving objects
  • Lucas-Canade method
  • Optical flows
Edit: header| contents| footer| sidebar