Department of Information Technology

Computer assisted image analysis I

1TD396, Fall 2016, 5 credits


  • To give an introduction to the processing and analysis of digital images.


Representation of images in computers. Image types. Sampling. Image coding and compression. Image processing and image restoration. Point operators. Fundamentals of frequency analysis. Histogram operations. Neighbourhood operators. Mathematical morphology. Segmentation. Shape analysis and feature extraction. Classification and decision theory. Examples of applications from research and industry.

Lectures and Labs

Date Time Place Title Teacher Chapters
24/10 10:15-12:00 1111 L1: Introduction Robin Strand 1, 2.1-2.4
25/10 10:15-12:00 1111 L2: Pointwise Operators Kristina Lidayová 2.6-2.6.4,3.1-3.3
28/10 13:15-17:00 2516D LAB 1 Damian Matuszewski, Tomas Wilkinson
2/11 13:15-15:00 1111 L3: Filtering 1 Robin Strand 2.5.1-2.5.2, 3.4-3.7, 5.3
4/11 13:15-15:00 1111 L4: Filtering 2 Anders Brun 4.2-4.10, 4.11.1, 4.11.3, 5.4
8/11 13:15-17:00 2516D LAB 2 Marine Astruc, Tomas Wilkinson
9/11 13:15-15:00 1111 L5: Mathematical Morphology and Distance Transforms Robin Strand 9.1-9.5.8, 9.6.1-9.6.3
15/11 10:15-12:00 1111 L6: Image Segmentation Robin Strand 10.1-10.2.5 and 10.3-10.5
18/11 13:15-15:00 1111 L7: Object Description Damian Matuszewski 11.1-11.4
22/11 13:15-17:00 2516D LAB 3 Kristina Lidayová, Tomas Wilkinson
28/11 13:15-15:00 1111 L8: Classification Marine Astruc
30/11 13:15-17:00 1549 LAB 4 Marine Astruc, Tomas Wilkinson
2/12 13:15-15:00 1111 L9: Color Images and Image Compression Anders Brun
8/12 13:15-15:00 1111 L10: Summary, old exams Robin Strand
12/12 13:15-17:00 2516D LAB 5 Damian Matuszewski, Tomas Wilkinson
19/12 14:00-19:00 Polacksbacken, Skrivsalen EXAM

Preliminary date for re-exam: April 18 2017


During the lab sessions, you will be able to try out things learned in the lectures. Please come prepared! That means, go over the lectures again and read the relevant sections in the book. If you are properly prepared, you should be able to finish all the exercises during the lab session.

If all your labs are done (and corrected) by the end of the semester (period 2) you will get your course credits this semester. Otherwise you will not get your credits until the reexam.

Lab 5 - CAPTCHA competition

Team Average Precision on test dataset [%] Running time [s]
tengil 91.52 18.41
FB 94.88 12.98
Iris 81.08 49.89
Josef 88.68 9.3
Team Norway 91.56 90.27
1200 if statements later 98.4 62.26
Last Fourier Transform 87.96 124.25
Lone Wolf 94.48 16.61
Better late than never 99.88 21.98
FCDW (Fastest Captcha Decoder in the West) 60.0 39.04
teamFischeezz 99.4 61.71
Iron Ladies 47.96 124.78
CAPTCHA Wars: The Course Awakens 89.12 106.95
CodingGuineaPig 89.6 24.86
Indigo 98.68 28.39


The labs are mandatory and for each lab finished before its deadline you will receive bonus points on the first question on the written exam (the first question gives max. 5 points and the exam will have a total of 40 points). If you cannot attend a lab, you are welcome to email a report to the lab assistants. Labs handed in during the course will be corrected as soon as possible.

On the written exam you will be allowed to carry one sheet (two sides) with hand-written notes. Preparing your own notes for the exam is a helpful studying technique, and we see little point in memorizing things.


Most of the lectures are based on this book:

  • R. C. Gonzalez and R. E. Woods, "Digital Image Processing",3rd. ed. : Upper Saddle River, N.J. : Prentice Hall, cop. 2008, ISBN: 978013505267-9.

You can get it (or order it) at e.g. Studentbokhandeln AB here in Uppsala or online, for example at: or A few books are also available through libraries.

Copies of slides and other material will be handed out during most of the lectures. The written exam is based on content covered in the lectures (oral presentation, lecture notes and blackboard examples). The book is highly recommended for background information and details about methods and theory.

Updated  2017-01-09 14:33:18 by Tomas Wilkinson.