Skip to content

Computer Vision

Public syllabus for 2025-2026

Back to syllabi

Academic overview

Programme
CS
Period
Year 1, Semester 1
Credits
5
Weeks
14

Curriculum placement

Appears in study plans

CS (2025 - 2027) Year 1, Semester 1 Elective subject, group 2

Teaching team

Course coordinator
Cosmin Bonchiș
Seminar coordinators
(none)

Learning time distribution

Total
Curriculum Lecture Practice Total Weekly Lecture Practice
42 28 14 3 2 1
Exam hours
6
Individual Study Bibliography study Field study Homework Tutoring Others
83 20 20 31 6 0
Overall
125

Learning outcomes

Knowledge

  • Acquisition of fundamental knowledge related to the basic techniques and methods in image processing and computer vision.

Skills

  • Ability to analyze images and extract information from them.
  • Ability to identify the appropriate algorithm/method for automatically analyzing images.

Responsibility

  • Capacity to independently solve specific tasks.
  • Ability to correctly and efficiently identify and plan the tasks specific to a given project.
  • Ability to responsibly assume professional duties and to comply with ethical and professional standards.
  • Ability to adapt to new requirements and modes of carrying out professional activities.

Online platform

(none)

Course content

Content Methods Obs
C1. Introduction to Computer Vision. Examples of digital image processing applications. Examples of computer vision applications. Digital image acquisition. Elements of linear algebra used in image processing. Basic relationships between pixels. Lecture, discussion, exemplification. 2h
C2. Image Processing in the Spatial Domain. Gray-level transformations. Histograms. Convolution. Edge detection. Gradient. Laplacian. Lecture, discussion, exemplification. 2h
C3. Image Processing in the Frequency Domain. The frequency domain and the Fourier Transform. Frequency-domain filtering: smoothing, sharpening, homomorphic filtering. Lecture, discussion, exemplification. 2h
C4. Segmentation. Thresholding. Edge-based segmentation. Hough Transform. Region-based segmentation. Lecture, discussion, exemplification. 2h
C5. Morphological Operations and Color Spaces. Dilation and erosion. Opening and closing. Skeletonization. Color spaces. Color transformations. Color-based segmentation. Lecture, discussion, exemplification. 2h
C6. Shape Representation and Description. Identification of regions of interest. Contour-based shape representation: chain codes. Region-based shape representation: scalar region descriptors, moments, convex hull. Lecture, discussion, exemplification. 2h
C7. Texture. Statistical texture descriptors: spatial frequency. Co-occurrence matrix. Edge frequency. Local Binary Patterns (LBP). Lecture, discussion, exemplification. 2h
C8. Object Recognition (1). Statistical pattern recognition: classifiers. Lecture, discussion, exemplification. 2h
C9. Object Recognition (2). Support Vector Machines (SVM). Histogram of Oriented Gradients (HOG). Lecture, discussion, exemplification. 2h
C10. Introduction to Convolutional Neural Networks (CNNs). Convolutional layer, max-pooling layer, activation functions, activation visualization. Lecture, discussion, exemplification. 2h
C11. Boosting Classifiers. The Viola–Jones algorithm for face detection. Lecture, discussion, exemplification. 2h
C12. Tracking. Mean-shift tracking. Cam-shift tracking. Lecture, discussion, exemplification. 2h
C13. Stereovision. Calibration. 2D–3D correspondence algorithms. Fundamental matrix. Lecture, discussion, exemplification. 2h
C14. Computer Vision in Everyday Life. Examples of solving real-world problems using computer vision. Lecture, discussion, exemplification. 2h

Course bibliography

Bibliografie [1] Gonzales R., Woods R., Digital Image Processing, 2nd/3rd edition [2] Sonka M., Hlavac V., Boyle R., Image Processing, Analysis and Machine Vision, 2nd edition [3] Forsyth, Ponce, Computer Vision A modern approach, 2nd edition.

Seminar content

Content Methods Obs
L1. Introduction to the PyCharm development environment. Python commands for matrix operations. Displaying images. Histograms. Convolution. Edge detection. Gradients. Problem-solving, dialogue, collaborative learning 2h
L2. Displaying images in the frequency domain. Frequency-domain filtering. Hough Transform. Problem-solving, dialogue, collaborative learning 2h
L3. Morphological operations. Conversion between color spaces. Problem-solving, dialogue, collaborative learning 2h
L4. Texture. Implementation of Gabor Filters. Implementation of Local Binary Patterns (LBP). Problem-solving, dialogue, collaborative learning 2h
L5. Implementation of the Histogram of Oriented Gradients (HOG) algorithm. Problem-solving, dialogue, collaborative learning 2h
L6. Running the Mean-shift Tracking algorithm. Building a Convolutional Neural Network (CNN). Problem-solving, dialogue, collaborative learning 2h
L7. Stereovision. Generating 3D images from two 2D images. Problem-solving, dialogue, collaborative learning 2h
Bibliography: Bibliografie1. J. E. Solem, Programming Computer Vision with Python, O’Reilly, 1st Edition, 2012 2. F. Cholet, Deep Learning with Python, Manning, 2015

Seminar bibliography

The content is consistent with the structure of similar courses offered at other universities and covers the fundamental aspects of applying computer vision techniques.

Corroboration

(none)

AI tools guidance

(none)

Evaluation and delivery

Activity Criteria Methods Percentage
C
  • Knowledge of computer vision concepts and techniques
  • Written examination
  • 30.0%
C
  • Correct identification of the appropriate image analysis method for solving a specific problem.
  • Project development
  • 40.0%
S
  • Use of suitable software tools and implementation of computer vision algorithms.
  • Laboratory applications and assignments
  • 30.0%

Performance standards

Attendance at lectures and seminars according to the general faculty requirements. Knowledge of the basic concepts of image processing and computer vision. Ability to implement a computer vision algorithm. Ability to identify the appropriate classification, clustering, or regression technique for solving a real-world problem. The final grade is calculated as a weighted average of the grades obtained for the components specified in the lecture and laboratory activities. The exam is considered passed if the average grade is at least 5, with each individual grade being no less than 5. Particular attention will be given to the work carried out throughout the semester, active participation in laboratory sessions, as well as during lectures — to ensure a better understanding of the concepts taught and a more accurate evaluation. In each examination session (including resit and grade improvement sessions), the grade will be calculated according to the same rule.

Additional info

(none)