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
|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|
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]|
|1200 if statements later||98.4||62.26|
|Last Fourier Transform||87.96||124.25|
|Better late than never||99.88||21.98|
|FCDW (Fastest Captcha Decoder in the West)||60.0||39.04|
|CAPTCHA Wars: The Course Awakens||89.12||106.95|
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.
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.