Course - Visual Informatics - IDIG4200
Visual Informatics
New from the academic year 2026/2027
Assessments and mandatory activities may be changed until September 20th.
About
About the course
Course content
This course introduces the fundamental principles and computational methods for visual information analysis and processing. It covers image enhancement, restoration, feature extraction, segmentation, and the basics of visual quality assessment. The course emphasizes both theoretical understanding and practical implementation using modern tools and datasets.
The course outline includes, but is not limited to:
- Introduction to Visual Informatics
- Overview of visual information systems and image data types
- Digital image representation, color models, and sampling
- Image Enhancement and Restoration
- Point and histogram-based enhancement techniques (contrast stretching, equalization)
- Denoising : spatial filters, frequency-domain filtering, etc.
- Deblurring : inverse filtering, Wiener filtering, blind deconvolution
- Dehazing and visibility restoration models
- Homomorphic filtering for illumination correction
- Feature Detection and Description
- Edge detection
- Corner and interest point detection
- Texture descriptors
- Shape descriptors
- Image and Video Segmentation
- Thresholding and region-based methods
- Clustering and/or graph-based segmentation
- Motion segmentation and optical flow
- Introduction to object detection and recognition
- Image and Video Quality Assessment (Basics)
- Subjective vs. objective quality
- Full-reference and no-reference metrics (PSNR, SSIM, etc.)
- Introduction to perceptual and learning-based quality assessment
Learning outcome
Knowledge
Upon completing the course, students will:
- Explain the physical and mathematical principles behind digital image formation and degradation.
- Explain key concepts in image enhancement, restoration, and feature extraction.
- Describe the main approaches for segmentation, detection, and recognition in visual data.
- Explain the fundamental principles of objective and subjective image quality assessment.
Skills
Upon completing the course, students will:
- Implement classical and modern methods for denoising, deblurring, and dehazing.
- Apply feature descriptors for edge, corner, texture, and shape analysis.
- Design and evaluate basic image/video segmentation and object detection pipelines.
- Perform quantitative quality evaluation of visual content.
General competence
Upon completing the course, students will:
- Critically assess algorithmic design choices in image processing systems.
- Communicate experimental results clearly using visual and quantitative analyses.
- Work effectively in small research-oriented teams on visual informatics tasks.
Learning methods and activities
- lectures
- Lab work
- Assignments
Mandatory requirements:
To be eligible for the final exam and project hand-in students are expected to deliver and get approved at least 80% of all the assignments during the semester
Compulsory assignments
- Compulsory assignments
Further on evaluation
See "Compulsory assignment" explained in Teaching Methods.
The student must obtain a passed grade in both two mandatory elements of assessment (the written exam and the projects) in order to complete the course.
The final project, its scope and the deadlines for the assignments and the project are announced during the semester. Students are expected to perform independent coding. Limited use of generative AI tools for learning purposes are accepted as long as the student documents its use and is able to understand and explain what they have delivered.
That is, programming and writing documentation by their own. Teaching assistants will be able to help students during the tutorial/lab session.
There will be a re-sit for the written exam in August. The re-sit examination can be oral.
The projects need to be resubmitted next time the course is run.
Specific conditions
Admission to a programme of study is required:
Informatics (MSIT)
Recommended previous knowledge
Basic linear algebra, probability, and programming (Python or MATLAB)
Course materials
Course book:
- Digital Image Processing, 4th Edition (DIP / 4th), by Rafael C. Gonzalez and Richard E. Woods, Prentice Hall (2017)
- Digital Image Processing Using MATLAB (DIPUM), by Rafael C. Gonzalez, Richard E. Woods, and Steven L. Eddins, Pearson (2018).
Further reading material:
- Color Image Processing: Methods and Applications (Image Processing), by Rastislav Lukac & Kostantinos N. Plataniotis, CRC (2006)
- The Image Processing Handbook, Fifth Edition (Image Processing Handbook), by John C. Russ, CRC (2006)
Subject areas
- Computer Science