Signal and Image Processing (SIP)

Course content

The course introduces basic computational, statistical, and mathematical techniques for representing, modeling, and analysing signals and images. Signals and images are measurements, which change with time and/or space, and these measurements typically originate from a physical system ordered on a grid. Examples are 1 dimensional sound, 2 dimensional images from a consumer camera, 3 dimensional reconstructions from medical scanners, and movies.

Applications include: removal of high frequency noise in sound, detecting and segmenting objects in images, and reconstruction of 3 dimensional Computed Tomography images (CT) from X-ray images.

Education

MSc programme in Computer Science

MSc programme in Physics

Learning outcome

Knowledge

  • Signal and image processing fundamentals
  • Sampling, Sampling theorem, Fourier transform
  • Convolution, linear and nonlinear filtering
  • Image restoration, inverse filtering
  • Image histograms
  • Image segmentation
  • Multiresolution processing
  • Radon transform
  • Linear and non-linear spatial transformaitons of images
  • Mathematical morphology

 

Skills

  • Apply basic signal processing methods to solve basic signal processing problems.


Competences

  • Evaluate which signal / image processing methods and pipeline of methods is best suited for solving a given signal problem.
  • Understand the implications of theoretical theorems and being able to analyse real problems on that basis.


 

The course will be a mixture of lectures, pen-and-paper exercises, and programming exercises.

See Absalon when the course is set up.

ECTS
7,5 ECTS
Type of assessment
Continuous assessment
Continuous evaluation of 7 written assignments.
Marking scale
7-point grading scale
Censorship form
No external censorship
Several internal examiners
Criteria for exam assessment

To get maximum grade the student must successfully be able to:

  • Discuss and apply the theoretical basics of digital signal and image processing

  • Reflect the linear processing of signals and design and apply digital filters for discrete signals.

  • Explain and identify different types of noise, design noise removal algorithms for image restoration and solve statistical linear inverse filtering problems for images.

  • Compare Fourier analysis to multiresolution analysis, relate the fundamental concepts of multiresolution analysis, and perform time/space-frequency analysis for signals and images.

  • Analyze the image histograms, and transform the images to another forms to enhance the visual content of the image or to facilitate easier interpretation and processing.

  • Explain fundamental image segmentation approaches and implement them to extract homogeneous regions on the images.

  • Relate and illustrate elementary representation methods in description of image content

 

Single subject courses (day)

  • Category
  • Hours
  • Lectures
  • 38
  • Theory exercises
  • 12
  • Practical exercises
  • 12
  • Preparation
  • 16
  • Project work
  • 128
  • English
  • 206