Course detail

Fundamentals of Image Processing

FEKT-MPA-FIPAcad. year: 2024/2025

At the beginning of the course, students will get acquainted with the theory of digitization of image data, their computer representation and data formats. The following is a description of the function of 2D operators for linear and nonlinear filtering of images with specific examples of their use. Subsequently, students will get acquainted with the basic methods of pattern and object recognition in images, image segmentation, object tracking, principles of stereoscopy and reconstruction of 3D objects. Finally, the basic principles of modern methods using machine learning (neural networks, deep learning for regression, classification and segmentation) will be explained.

Language of instruction

English

Number of ECTS credits

5

Mode of study

Not applicable.

Entry knowledge

Knowledge of mathematics (derivatives, integrals, matrix and vector calculus), basics of programming and basics of signal processing is required. 

Rules for evaluation and completion of the course

The evaluation of the course will take place in the framework of practical computer exercises, continuous tests, projects and a final exam. The exact point evaluation of the course and the method of its implementation is determined by the annually updated decree of the subject guarantor. The final exam is realized in combined form. 
The definition of controlled teaching and the method of its implementation is determined by the annually updated decree of the subject guarantor.

Aims

The aim of the course is to acquaint students with the basic methods of processing and analysis of digital image data, basic operations with images and concepts of modern methods used in autonomous vehicle control, to show examples and basic functionality of these methods on practical demonstrations. 
The graduate of the course is able to: (a) explain the principle and procedure of image digitalization, (b) explain the principles of 2D digital systems, (c) perform basic digital image processing operations, (d) analyze the basic properties and information contained in digital images, (e) to orientate in the application of modern methods of image processing and analysis using machine learning. 

Study aids

Not applicable.

Prerequisites and corequisites

Not applicable.

Basic literature

JAN, J. Medical image processing reconstruction and analysis: concepts and methods. Second edition. Boca Raton: CRC Press, 2019. ISBN 978-113-8310-285. (EN)

Recommended reading

WALEK, P., LAMOŠ, M a JAN, J. Analýza biomedicínských obrazů: počítačová cvičení. 2. Brno: Vysoké učení technické v Brně, FEKT, ÚBMI, 2015. ISBN 978-80-214-4792-9. (CS)
JAN, J. Číslicová filtrace, analýza a restaurace signálů. 2. upr. a rozš. vyd. Brno: VUTIUM, 2002. ISBN 80-214-1558-4. (CS)
JAN, J. Digital signal filtering, analysis and restoration. London: Institution of Electrical Engineers, 2000. IEE telecommunications series, 44. ISBN 978-085-2967-607. (EN)

Elearning

Classification of course in study plans

  • Programme MPA-AEE Master's 1 year of study, winter semester, compulsory-optional

Type of course unit

 

Lecture

26 hod., optionally

Teacher / Lecturer

Syllabus

1. Basics of signal representation of images and their spectra

2. Digital images and digital operators

3. Basic methods of modifying images with point operators

4. Basic methods of image editing with local operators in the spatial and frequency domain

5. Parametric images and texture analysis

6. Methods for detecting and tracking objects in images

7. Image segmentation methods

8. Geometric transformations of images

9. Stereoscopy and its use for distance estimation

10. Methods of 3D object reconstruction using stereoscopy and multiscopy

11. Machine learning methods for classification and regression

12. Principles of deep learning methods and convolutional neural networks

13. Architectures and applications of deep learning methods in autonomous driving

Exercise in computer lab

26 hod., optionally

Teacher / Lecturer

Syllabus

1. Basics of signal representation of images and their spectra

2. Digital images and digital operators

3. Basic methods of modifying images with point operators

4. Basic methods of image editing with local operators in the spatial and frequency domain

5. Parametric images and texture analysis

6. Methods for detecting and tracking objects in images

7. Image segmentation methods

8. Geometric transformations of images

9. Stereoscopy and its use for distance estimation

10. Methods of 3D object reconstruction using stereoscopy and multiscopy

11. Machine learning methods for classification and regression

12. Principles of deep learning methods and convolutional neural networks

13. Architectures and applications of deep learning methods in autonomous driving

Elearning