Image Processing and Computer Vision (ELE510)

Content

The subject consists of different topics from image processing and computer vision, with the following content:

Introduction to image processing, representation of digital images, color, light,geometric and point transformations, binary image processing, spatial filtering, frequency domain filtering, noise removal in images, edge detection and corner/feature detection, image segmentation, feature extraction, texture properties, image formation, perspective projection, geometric camera model, camera calibration, stereopsis, recording with two cameras.

Finally, some topics on deep neural networks in image processing.

Learning outcome

Knowledge:

At the end of the semester a successful student should have knowledge about the following topics:

1. Representation of digital images, including basic knowledge of light, shadow and color.
2. Linear and non-filtering of images. Filters can be used for noise removal, edge detection and analysis of image properties.
3. Extracting features, for example texture features and corners from images
4. Denoising of images
5. Segmentation of images
6. Principles of geometric camera models.
7. Imaging with several cameras, especially stereo vision.
8. Know the principles of segmentation, grouping and modeling in image processing and computer vision.
9. Have som knowledge on the use of deep neural networks in image processing tasks.

Skills:

At the end of the semester a successful student should have skills in processing and analysis of digital images and be able to design simple robot vision systems and machine vision systems. The student should have skills in using python ( labs given as jupyter notebooks) and openCV for the purpose of image processing and image analysis.

Competence:

After this course the student should have a general understanding of fundamental image processing and computer vision, where some concepts will be known at a more superficial level than others. The student should have a general understanding of how to use image processing and computer vision methods in real world applications.

Required prerequisite knowledge

Recommended prerequisites

To be able to take this course, you are expected to have a bachelor's degree in engineering or technology. It is expected that you have had a course in programming as Python programming must be used in mandatory submissions. Mathematical knowledge equivalent to the completion of a bachelor's program in engineering is expected. More specifically, knowledge of linear algebra with matrices and vector calculus, Fourier series and complex numbers is expected. It is also expected that you understand and can use concepts from probability and statistical methods such as probability density functions, expected values, variance and median.

It is an advantage, but not a requirement, to have knowledge of sampling and digitization of signals, Fourier transform, and discrete Fourier transform.

Exam

Coursework requirements

Mandatory exercises: A set of mandatory exercises/laboratory work must be approved.

Mandatory work demands (such as hand in assignments, lab- assignments, projects, etc) must be approved by subject teacher within the specified deadlines.

Compulsory course attendance that must be completed and approved before access to the laboratory: Electronic Course in Health, Safety and Environment.

Completion of mandatory lab assignments are to be made at the times and in the groups that are assigned and published. Absence due to illness or for other reasons must be communicated as soon as possible to the laboratory personnel. One cannot expect that provisions for completion of the lab assignments at other times are made unless prior arrangements with the laboratory personnel have been agreed upon.

Failure to complete the assigned labs on time or not having them approved will result in barring from taking the exam of the course.

Course teacher(s)

Course teacher:

Coordinator laboratory exercises:

Course coordinator:

Head of Department:

Method of work

4 hours lectures a week. Video lectures and Q&A classes can be used instead of ordinary lectures on some parts of the material.

Allotted time for exercises with a teaching assistant (TA) present: 2 hours per week.

Open for

Course assessment

Literature