ID:
509500
Durata (ore):
28
CFU:
3
SSD:
SISTEMI DI ELABORAZIONE DELLE INFORMAZIONI
Anno:
2026
Dati Generali
Periodo di attivitÃ
Secondo Semestre (01/03/2027 - 11/06/2027)
Syllabus
Obiettivi Formativi
- Knowledge and understanding: Students will understand the physical and mathematical foundations of digital image acquisition, the mechanics of human visual perception, and the theoretical principles of image processing (IP) and computer vision (CV)
- Applying knowledge and understanding: Students will be able to design and implement algorithms in Python for image enhancement, spatial filtering, and color space manipulation. They will be capable of building basic machine learning pipelines for segmentation and classification.
- Making judgements: Students will learn to evaluate the appropriateness of traditional image processing techniques versus modern deep learning approaches for specific engineering problems.
- Learning skills: Students will develop the ability to autonomously integrate theoretical concepts with practical software implementation through coding assignments.
- Applying knowledge and understanding: Students will be able to design and implement algorithms in Python for image enhancement, spatial filtering, and color space manipulation. They will be capable of building basic machine learning pipelines for segmentation and classification.
- Making judgements: Students will learn to evaluate the appropriateness of traditional image processing techniques versus modern deep learning approaches for specific engineering problems.
- Learning skills: Students will develop the ability to autonomously integrate theoretical concepts with practical software implementation through coding assignments.
Prerequisiti
Students should possess basic introductory notions of Artificial Intelligence and Mathematics from the previous mandatory courses.
Exercises are conducted in the Python programming language, for which a basic knowledge is advised.
Exercises are conducted in the Python programming language, for which a basic knowledge is advised.
Metodi didattici
Frontal Lectures (16 hours, 2 CFU): Conducted with slides to illustrate complex visual concepts, such as image formation and filtering effects.
Practical Exercises (12 hours, 1 CFU): Guided Python-based sessions where students solve image processing problems, ensuring they can apply theoretical models to real digital data.
Assignments: Non-mandatory Python assignments to encourage individual problem-solving and deeper exploration of selected topics.
Practical Exercises (12 hours, 1 CFU): Guided Python-based sessions where students solve image processing problems, ensuring they can apply theoretical models to real digital data.
Assignments: Non-mandatory Python assignments to encourage individual problem-solving and deeper exploration of selected topics.
Verifica Apprendimento
- Written Test: A mandatory exam featuring three types of questions:
-- Open-ended questions: Focused on theoretical descriptions and the ability to use drawings or schemes.
-- Exercises: Practical problems requiring mathematical expressions and graphical explanations.
-- Closed-ended questions: Design-oriented scenarios where students must select and motivate the best tools for a specific task.
- Python Assignments (Optional): Three individual assignments contribute up to 1.5 additional points (0.5 each) to the final grade.
Scoring:
The exam is graded on a scale of 0-30. The minimum passing score is 18 in the written test.
A perfect score in the written test plus all assignments results in a 31.5 (30 cum laude) for MOD.1
Integration:
MOD.1 (Image Processing) is part of the "Signal and Image Processing" integrated course.
The final grade for the entire course will be the arithmetic average of the grades obtained in MOD.1 and MOD.2.
-- Open-ended questions: Focused on theoretical descriptions and the ability to use drawings or schemes.
-- Exercises: Practical problems requiring mathematical expressions and graphical explanations.
-- Closed-ended questions: Design-oriented scenarios where students must select and motivate the best tools for a specific task.
- Python Assignments (Optional): Three individual assignments contribute up to 1.5 additional points (0.5 each) to the final grade.
Scoring:
The exam is graded on a scale of 0-30. The minimum passing score is 18 in the written test.
A perfect score in the written test plus all assignments results in a 31.5 (30 cum laude) for MOD.1
Integration:
MOD.1 (Image Processing) is part of the "Signal and Image Processing" integrated course.
The final grade for the entire course will be the arithmetic average of the grades obtained in MOD.1 and MOD.2.
Testi
Gonzalez, R. C., & Woods, R. E. (2018). Digital Image Processing. 4th Edition. Pearson. ISBN-13: 978-0133356724.
(http://www.imageprocessingplace.com/index.htm)
Any other versions of the book is acceptable.
Additional teaching materials, including lecture slides and exercise notebooks, will be provided via the e-learning platform.
(http://www.imageprocessingplace.com/index.htm)
Any other versions of the book is acceptable.
Additional teaching materials, including lecture slides and exercise notebooks, will be provided via the e-learning platform.
Contenuti
The course is structured around six main topics:
- Fundamentals: Geometry of vision, color vision, image digitization (sampling, quantization, aliasing), and the camera imaging pipeline.
- Point-wise and Global Operators: Linear and non-linear intensity transformations (e.g., gamma correction, contrast enhancement) and histogram equalization.
- Local Operators: Noise types (Gaussian, Impulse), spatial filtering (blurring, sharpening, edges), and mathematical morphology (erosion, dilation, opening, closing).
- Color Imaging: Color spaces (RGB, sRGB, HSL, HSV), color image processing, and computational color constancy (white balance).
- Machine Learning for Computer Vision: Segmentation techniques (binary masks, dense partitioning), feature extraction (color, shape, texture/Gabor filters), and decision-making (classification, regression).
- Modern Outlook: A comparison between traditional Machine Learning and Deep Learning, identifying when to use each approach.
- Fundamentals: Geometry of vision, color vision, image digitization (sampling, quantization, aliasing), and the camera imaging pipeline.
- Point-wise and Global Operators: Linear and non-linear intensity transformations (e.g., gamma correction, contrast enhancement) and histogram equalization.
- Local Operators: Noise types (Gaussian, Impulse), spatial filtering (blurring, sharpening, edges), and mathematical morphology (erosion, dilation, opening, closing).
- Color Imaging: Color spaces (RGB, sRGB, HSL, HSV), color image processing, and computational color constancy (white balance).
- Machine Learning for Computer Vision: Segmentation techniques (binary masks, dense partitioning), feature extraction (color, shape, texture/Gabor filters), and decision-making (classification, regression).
- Modern Outlook: A comparison between traditional Machine Learning and Deep Learning, identifying when to use each approach.
Lingua Insegnamento
INGLESE
Corsi
Corsi
ARTIFICIAL INTELLIGENCE
Laurea
3 anni
No Results Found
Persone
Persone
Docente
No Results Found