Course details

Student Mobility > Programmes and Courses > Courses in English > Course details

Machine Vision Algorithms

Teaching: Completely taught in English

ECTS: 6

Level: Graduate

Semester: Summer

Prerequisites:

Load:

Lectures	Exercises	Laboratory exercises	Project laboratory	Physical education excercises	Field exercises	Seminar	Design exercises	Practicum
30		30	0	0	0	15	0	0

Course objectives:

Mastering the knowledge of application and programming of machine vision algorithms. Solving practical problems in the field of machine vision.

Student responsibilities:

Class attendance. Independent work (seminar). Written exam.

Grading and evaluation of student work over the course of instruction and at a final exam:

Independent work (seminar). Written exam.

Upon successful completion of the course, students will be able to (learning outcomes):

1 . Explain the principles of operation and define the concepts related to machine vision

2 . Systematically approach the analysis and solution of problems that require the application of machine vision algorithms

3 . Use the acquired knowledge for the development of technical systems with integrated vision systems

4 . Independently apply or develop modifications to basic machine vision algorithms for a specific task

5 . Analyze and critically evaluate the quality of machine vision application

Lectures

1. The role and tasks of machine vision algorithms.

2. Perspective transformations, image formation, model and calibration of digital cameras.

3. Image processing I: pixel operations, color transformations, histogram.

4. Image processing II: linear filters, nonlinear filters, Fourier, morphology.

5. Segmentation: thresholding methods.

6. Edge detection: Gaussian operator, Sobel, Canny, LoG operator (Laplacian of Gaussian), Marr-Hildreth, Roberts, Prewitt, Kirsch, Robinson.

7. A systematic approach to problem solving using machine vision algorithms. Solving tasks.

8. Recognition of shapes, Hough transforms, recognition of circles and ellipses.

9. Localization of features and objects, detector features.

10. Methods of registering and finding correspondence between features of interest.

11. The role of machine learning in solving problems of recognition and classification.

12. Algorithms for processing 3D data structure (point cloud).

13. Algorithms for spatial reconstruction of scenes and objects.

14. Case study - visual control of products, equipment, algorithms.

15. Case study - identification and localization of objects.

Exercises

1. Introduction to software tools.

2. Representation of digital image, image processing, basic functions.

3. Pixel operations, color transformations, histogram analysis.

4. Linear filters, nonlinear filters, Fourier, morphology.

5. Segmentation: threshold determination methods.

6. Boundary segmentation.

7. Localization of objects.

8. Development and application of shape recognition algorithm.

9. Localization of objects, template correspondence.

10. Methods of registration and finding correspondence.

11. Solving a practical task.

12. Solving a practical task.

13. Solving a practical task.

14. Solving a practical task.

15. Solving a practical task.

Compulsory literature:

1. Computer and machine vision: theory, algorithms, practicalities, E. R. Davies, Boston: Elsevier, 2012, p. 0-0

2. Computer vision: algorithms and applications, Szeliski, Richard, Springer Science & Business Media, 2010, 2010, p. 0-0

Recommended literature: