Course detail

Numerical Methods of Image Analysis

FSI-TNMAcad. year: 2023/2024

The course familiarises students with the digital image processing theory and selected topics of image analysis. It focuses on digital image representation and reconstruction, filtration in frequency and spatial domain, noise analysis and filtration, image enhancement, image segmentation, objects analysis and recognition, analysis of multi-spectral images.

Language of instruction

Czech

Number of ECTS credits

4

Guarantor

prof. RNDr. Miloslav Druckmüller, CSc.

Department

Institute of Mathematics (ÚM)

Entry knowledge

Real and complex analysis, functional analysis, basic knowledge of programming

Rules for evaluation and completion of the course

Course-unit credit based on a written test.
Exam has a written and an oral part.


Missed lessons can be compensated via aditional exercises.

Aims

The aim of the course is to provide students with information about modern mathematical methods of image processing, including programming techniques.


Basic knowledge of classic and digital photography, modern mathematical methods of image processing,
image analysis and pattern recognition.

Study aids

Not applicable.

Prerequisites and corequisites

Not applicable.

Basic literature

Pratt, W. K.: Digital Image Processing (Second Edition), New York: Wiley 1991 (EN)
Petrou, M., and Bosdogianni, P., Image Processing: The Fundamentals, John Wiley& Sons, 1999. (EN)

FOLLAND, Gerald B, 1992. Fourier analysis and its applications. Providence: American Mathematical Society, 433 s. Pure and applied undergraduate texts. ISBN 978-0-8218-4790-9.

(EN)

BEZVODA, Václav, Josef JEŽEK, Stanislav SAIC a Karel SEGETH, 1988. Dvojrozměrná diskrétní Fourierova transformace a její použití: I. Teorie a obecné užití. 1. Praha: Státní pedagogické nakladatelství, n. p., 181 s. ISBN 17-135-88.

(CS)

Recommended reading

Klíma, M.; Bernas, M.; Hozman J.; Dvořák, P. : Zpracování obrazové informace, , 0 (CS)
Druckmüller, M.; Heriban, P.: Digital Image Processing System 5.0, , 0 (CS)

Type of course unit

 

Lecture

26 hours, optionally

Teacher / Lecturer

prof. RNDr. Miloslav Druckmüller, CSc.

Syllabus

1. Principles of classic and digital photography
2. Numeric image representation, graphics formats, image data compression
3. Images reconstruction, statistical image characteristics
4. Pixel values transforms
5. Convolution, space domain filtration
6. Fourier transform, frequency domain filtration
7. Low-pass and high-pass filters, nonlinear filters
8. Adaptive filters
9. Additive noise - analysis and filtration
10. Impulse noise - analysis and filtration
11. Image segmentation
12. Object analysis
13. Pattern recognition and object classification

Computer-assisted exercise

26 hours, compulsory

Teacher / Lecturer

prof. RNDr. Miloslav Druckmüller, CSc.

Syllabus

1. Color spaces, histogram
2. Image visualization - pixel value transformation
3. Fourier transform of functions of two real variables and its properties
4. One-dimensional and two-dimensional discrete Fourier transform
5. Visualization of spectrum, its basic modifications
6. Convolution, image filtration, additive noise
7. Impulse noise
8. Analysis of directions in image
9. Image registration, phase correlation
10. Image compression, cosine transform, JPG
11. Image segmentation - thresholding, edge detection
12. Object analysis, moment method
13. Reserve of the lecturer, semestral project