ITD33517 Image Processing and Pattern Recognition (Spring 2018)

The course is connected to the following study programs

Elective course.

Prerequisites

This course requires prerequisite knowledge in:

• mathematics equivalent to the course in Mathematics for Computer Science
• programming equivalent to the course in Object-oriented Programming

The student's learning outcomes after completing the course

Knowledge:

On completion of the course, the student:

• understands digital images and their properties
• is familiar with representation methods for digital images, including image coding and compression
• understands how images can be affected by noise
• understands how images can be filtered/processed for better quality
• understands the principles and methods in basic pattern recognition

Skills:

On completion of the course, the student can:

• use standard filters to improve image quality by filtering noise
• use standard filters to enhance the details in an image
• use standard techniques to detect edges, corners and objects in an image
• write programs for basic image processing and pattern recognition
• get a computer and/or computer system to recognize objects in an image

Content

Representation methods for digital images, image coding and compression, noise reduction techniques, use of filters, detail enhancement, edge detection, image transformations, handling shape in an image, pattern recognition and classification and programming methods for image processing.

Forms of teaching and learning

Lectures, exercises and lab assignments/projects.

If students from international partners attend courses, the lectures will be conducted in English.

Workload

4 hours of lectures per week, as well as study groups and lab assignments/projects.

Coursework requirements - conditions for taking the exam

Up to four compulsory assignments.

The coursework requirements must be approved before students may sit the exam.

Examination

Individual portfolio assignment

The portfolio will include the following:

• a self-developed program in line with the syllabus
• a report/article (approximately 6-12 pages) describing the program/project

The portfolio is assessed as a whole and one overall grade is awarded.

Grading scale from A to F are used. In the event of a resit or rescheduled examination, all parts of the portfolio must be re-submitted.

Course evaluation

This course is evaluated as follows:

• Mid-semester evaluation (optional)
• Final evaluation (compulsory)

The course instructor prepares a course report on the basis of student feedback and on his/her own experiences of the course. Course reports are discussed with the Committee for Study Quality at the Faculty of Computer Sciences.

Literature

The literature list was last updated November 16, 2017

Web-based resources.

Recommended literature:
Gonzalez, R. C og Woods, R. E: Digital Image Processing. Pearson. 3. utgave. ISBN 978-0131687288.