Signal and image processing (2015/2016)

Course code
Gloria Menegaz
Other available courses
Teaching is organised as follows:
Unit Credits Academic sector Period Academic staff
SEGNALI 6 INF/01-INFORMATICS I semestre Paolo Fiorini
IMMAGINI 6 INF/01-INFORMATICS II semestre Gloria Menegaz

Learning outcomes

The course is meant to the acquisition of the main theoretical and practical tools that are necessary for image processing, including both natural and medical images.
It is recommended to be familiar with the main concepts that are introduced in the Signal part, in particular Fourier (time/frequency) analysis, linear time-invariant systems and digital filtering.
The whole pipeline from image acquisition to image rendering will be considered, and the main theorical and practical aspects will be treated.
Laboratory sessions will be organized in order to "put in practice" what is treated in the course covering the entire pipeline in order to provide the student a complete view on the pipeline as well as good desing and implementation capabilities.


- Introduction to image processing
- The acquisition pipeline (sampling, quantization)
- Image enhancement (in both spatial and frequency domain)
- Morphological image processing
- Color imaging
- Image segmentation (edge-based, region-based)
- Elements of pattern recognition

Assessment methods and criteria

Theory: written or oral exam (depending on the number of students)
- evaluation of the lab. sessions during the course
- Lab. specific questions during the exam
- Project only in case of diploma project as an alternative to questions

Statistics about transparency requirements (Attuazione Art. 2 del D.M. 31/10/2007, n. 544)

Outcomes Exams Outcomes Percentages Average Standard Deviation
Positive 100.0% 27 2
Rejected --
Absent --
Ritirati --
Canceled --
Distribuzione degli esiti positivi
18 19 20 21 22 23 24 25 26 27 28 29 30 30 e Lode
0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 20.0% 20.0% 0.0% 0.0% 0.0% 60.0% 0.0% 0.0%

Data from AA 2015/2016 based on 5 students. I valori in percentuale sono arrotondati al numero intero più vicino.