lunduniversity.lu.se

AI Lund

An open network for research, education and innovation in the area of Artificial Intelligence at Lund University

Denna sida på svenska This page in English

AI Lund Events

Image Analysis (7,5 ECTS) autumn 2019

Segmentation

Föreläsning

From: 2019-11-06 09:00
Place: To Be Decided
Contact: kalle [at] maths [dot] lth [dot] se
Save event to your calendar


Welcome to this course open both to PhD-students (free) as well as to professionals (fee). The aim of the course is to give necessary knowledge of digital image analysis for further research within the area and to be able to use digital image analysis within other research areas such as computer graphics, image coding, video coding and industrial image processing problems. 

Prerequisites 

Linear algebra and analysis in one and several variables. Highly practiced skills in conducting experiments, carrying out work in project form and in programming.

Schedule

Participants are assumed to have an interest in Image Analysis in their respective work. Participants should preferably bring one image analysis problem and data to the class to work on during the course and present progress on this in a presentation on January 15. For feepaying professionals it is also possible to just follow the lectures, but then no credits will be given.

Almost full day (9-15) of lectures on: 

  • November 6 
  • November 20
  • December 4
  • Optionally - present your project work on January 15 at 13:15. 

Work on your own image analysis problem in between and after.

Programme

Course coordinator

Kalle Åström, Professor in Mathematics, Deputy Head of Department at Centre for Mathematical Sciences, Lund University.

Knowledge and understanding

For a passing grade the participant must
● be able to explain clearly, and to independently use, basic mathematical concepts in image analysis, in particular regarding transform theory (in space as well as in the frequency domain), image enhancement methods, image compression and pattern recognition.
● be able to describe and give an informal explanation of the mathematical theory behind some central image processing algorithms (both deterministic and stochastic).
● have an understanding of the statistical principles used in machine learning.

Competences and skills

For a passing grade the participant must
● in an engineering manner be able to use computer packages to solve problems in image analysis.
● be able to independently apply basic methods in image processing to problems which are relevant in industrial applications or research.
● with proper terminology, in a well-structured manner and with clear logic be able to explain the solution to a problem in image analysis.

Contents

Basic mathematical concepts: Image transforms, Discrete Fourier Transform, Fast Fourier Transform.
Image enhancement: Grey level transforms, filtering.
Image restoration: Filterings, inverse methods.
Scale space theory: Continuous versus discrete theory, interpolation.
Extraction of special features: Filtering, edge and corner detection.
Segmentation: graph-methods, active contours, mathematical morphology.
Bayesian image handling: Maximum A Posterori (MAP) estimations, simulation.
Pattern recognition: Classification, SVM (Support Vector Machine), Principal Component Analysis (PCA), learning, deep learning.
Registration 

Machine Learning: Training, testing, generalization, hypothesis spaces.

Application

  • For PhD-students (free of charge) - kalle [at] maths [dot] lth [dot] se (subject: Registration%20%22Image%20analysis%22%20for%20doctoral%20students%20autumn%202019) (Register here to Kalle Åström!)
  • For Professionals (9,950 SEK excl VAT) - kimmo [dot] wallenius [at] fsi [dot] lu [dot] se (subject: Registration%20%22Image%20Analysis%22%20autumn%202019) (Register here to Kimmo Wallenius!)

Related links

https://fukurser.lth.se/fud/details/?code=FMA105F

The course is similar to FMAN20, 
https://canvas.education.lu.se/courses/1619