Past courses

Lecturer: Prof. Marco Di Felice, University of Bologna (marco.difelice3@unibo.it)

Duration: 4 classes of 3 hours each

Offered: June 2023

Abstract:
The course introduces techniques and tools for the design of data analytics processes for IoT applications through the Python language. By combining theoretical and practical contributions, it reviews the main stages of an IoT data pipeline: i.e. data acquisition, data visualization, data pre-processing, knowledge extraction through Machine Learning (ML) algorithms. After a brief recap of basic Python programming concepts, it presents the most used libraries (Pandas, Numpy) for the processing of tabular and vectorial data and for the 2D plotting (Seaborn, Matplotlib). Then, it introduces selected data processing and learning techniques for IoT time-series classification/forecasting and shows their implementations through Python frameworks (scikit-learn, Keras). Datasets and case studies related to IoT-based monitoring applications (SHM and smart agriculture) will be considered.

Final exam: a project

Lecturer: Prof. Enrico Vitucci, University of Bologna (enricomaria.vitucci@unibo.it)

Duration: 6 hours

Offered: 2023

Abstract: The course presents the fundamental concepts of remote sensing based on electromagnetic techniques, and its application to climate change observation, environmental monitoring, detection of air pollution, and sustainable urban planning, either using satellite, aerial, or terrestrial platforms. Different technologies are described, such as passive optical and microwave sensors, and active sensing systems such as Synthetic Aperture Radar and Lidar. Finally, ray-based techniques to evaluate the intensity of solar energy and its impact on the Urban Heat Island effect and on energy consumption in urban areas are outlined. Possible strategies to cope with Urban Heat Island and to increase the overall energetic efficiency of urban areas are discussed in the final part of the course.

Short course by Prof. Francesco Tinti promoted within the activities of our PhD program.

DATE: 03 February - 11 February 2022 from 16:00 to 18:00

EVENT LOCATION: Aula Laboratorio Informatico LAB6, via Terracini n. 28 – Bologna
In presence and online event

SCHEDULE OF THE COURSE:
03.02 h 16:00-18:00 - Regression
04.02 h 16:00-18:00 - Interpolation: linear estimator
10.02 h 16:00-18:00 - Interpolation: kriging
11.02 h 16:00-18:00 - Predictions and risks map
Each lecture will consist of a theoretical part and a practical application on excel.

For further information contact alessandro.marzani@unibo.it or francesco.tinti@unibo.it

Organized by the Image Processing Group of the Unione Matematici Italiani, for further information contact Prof. Serena Morigi (serena.morigi@unibo.it)

Duration: 24 hours

Offered: January/February 2022, free of charge

Structure:
Course composed of four blocks taught each in an Italian institution (UnInsubria, UniMoRe, UniBo, UniGe).
The course will take place in a blended format where, for each block, some local audience will attend classes “in presence” while the non-local students will follow the course via online platforms.

Course contents (roughly):
- 1st block: introduction to inverse problems in imaging, singular value decomposition, regularisation theory, Tikhonov filters, iterative regularization methods, structured matrices. Fourier transform in discrete setting (DFT, FFT), wavelet representation of signals, sparsity.
- 2nd block: numerical optimization methods in imaging: forward-backward methods in differentiable and non-differentiable optimization, variable metrics, inertial techniques, inexact solution of proximal operator, non-convex problems.
- 3rd block: ADMM, application to applied inverse problems in Imaging; restoration, reconstruction, segmentation, inpainting.
- 4th block: basic notions of statistical learning, supervised/unsupervised approaches, applications to standard imaging problems (classification..), deep learning.

Instructor: Dr. Nicola Paltrinieri, associate professor at NTNU Trondheim and Visiting fellow at the Institute of Advanced Studies of University of Bologna (nicola.paltrinieri@ntnu.no)

Duration: 8 hours

Offered: 14, 21, 28 October and 4 November 2021

Abstract:
Industry is stepping into its 4.0 phase by implementing and increasingly relaying on cyber-technological systems. Wider networks of sensors may allow for continuous monitoring of industrial process conditions. Enhanced computational power provides the capability of processing the collected “big data”. Early warnings can then be picked and lead to suggestion for proactive safety strategies, or directly initiate the action of autonomous actuators ensuring the required level of system safety. But have we reached these safety 4.0 promises yet, or will we ever reach them?
A traditional view on safety defines it as the absence of accidents and incidents. A forward-looking perspective on safety affirms that it involves ensuring that “as many things as possible go right”. However, in both the views there is an element of uncertainty associated to the prediction of future risks and, more subtle, to the capability of possessing all the necessary information for such prediction. This uncertainty does not simply disappear once we apply advanced artificial intelligence (AI) techniques, but it can be found behind modelling choices and parameters setting.
These seminars will illustrate a series of examples where AI techniques are used to continuously update the evaluation of the safety level in an industrial system. This will allow us to affirm that such advanced techniques are progressively providing a reliable support for critical decision making and guiding industry towards more risk-informed and safety-responsible planning.

Instructor: Prof. Marco Di Felice, University of Bologna (marco.difelice3@unibo.it)

Duration: 8 hours

Offered: July or September 2021, students of the XXXVI cycle

Abstract:
Location-aware systems denote a wide-range of ICT solutions able to acquire, visualize and process spatial data. At today, location can be considered the main enabler of a large number of high-value applications on different contexts, from health-care to logistics, from proximity marketing to the Internet of Things (IoT). This is explainable, from one side, by the capillar diffusion of Global Navigation Satellite Systems (GNSS) devices, with an installed base of more than 7 billion devices in 2020 worldwide (source: GSA); on the other side, the wide availability of data with spatial referencing has fueled the research on novel technologies for spatial data management, storing, mapping and analytics. The course provides an introduction to location-aware systems, from the enabling technologies to the use-cases, mainly focusing on applications related to IoT and IIoT monitoring. Specifically, each lesson will discuss a specific component of the context-aware cycle: i.e. (i) position data acquisition, by considering state-of-art positioning technologies for outdoor/indoor scenarios; (ii) position data modeling and storing, by introducing spatial data formats (e.g. GeoJSON, GeoTIFF) and spatial database management systems; (iii) position data visualization, by describing state-of-art map frameworks and APIs (e.g. OpenStreetMap) available on the market; (iiii) position data analytics, by introducing spatial correlation and data regression concepts. In addition, each lesson will feature demo and classroom exercises.

Instructors: Prof. Alessandro Marzani, University of Bologna (alessandro.marzani@unibo.it) and Dr. Antonio Palermo,, University of Bologna (antonio.palermo6@unibo.it)

Duration: 3 hours

Offered: July 21, 2021 9:30 AM

Abstract:
Many natural and engineered materials across different length scales present a geometry or a material composition with a periodic pattern. Originally, periodicity in both structural systems and engineered materials was introduced to achieve enhanced mechanical properties at a reduced weight. More recently periodicity has been utilized to design materials, structures or devices able to control the propagation of mechanical vibrations and waves. Periodic structures present, in fact, a unique dynamic response that can be exploited to impede the propagation of waves within desired frequency ranges and/or spatial directions. This course aims at introducing the fundamental aspects of elastic wave propagation in periodic structures. Basic concepts of lattice dynamics and wave propagation will be introduced by describing the response of 1D and 2D discrete systems. Dispersive features characteristic of periodic systems like bandgaps and wave beaming will be discussed. Additionally, appropriate computational techniques to extract the band structures properties of continuos periodic systems will be presented. In particular, the course will focus on Finite Element based techniques for band structure calculation, as the Wave Finite Element method and the Bloch operator Finite Element method.

Instructor: Prof. Luca De Marchi, University of Bologna (l.demarchi@unibo.it)

Duration: 8 hours

Offered: June 2021, students of the XXXVI cycle

Abstract:
The use of ultrasonic guided waves (GWs) for structural health monitoring (SHM) has interested many researchers. Nevertheless, Lamb-wave testing for SHM is complicated by the dispersive nature of wave modes, which deteriorates the wave spatial resolution and makes the experimental data hard to interpret. To tackle this problem, methods which analyze the dispersive signals in the domain spanned by time-frequency representations (TFRs) have been proposed. An infinite number of TFRs and processing tools can be obtained by using unitary transformations. Unitary operators such as the Warped Frequency Transform (WFT) are particularly suited for the analysis of GWs. This seminar will focus on WFT-based analysis methods capable to achieve sparse representations of GW signals. These methods naturally lead to super-resolvedand artifact-free representations, even in noisy environments, and are particularly effective to extract the information on the wave distance of propagation. The concept of sparse representations is also the basis of the so-called compressive sensing (CS) theory, which offers an intriguing alternative with respect to the classical process of acquiring signals according to the Shannon–Nyquist paradigm. CS theory proves that a signal which is sparse in a given representation can be compressed directly at the sampling stage. In this talk, a CS framework for Lamb wave field acquisitions with air-coupled probes or laser-Doppler vibrometers will be presented. The proposed framework is intended to minimize the number of ultrasonic scan point locations over the surface of the inspected structure. The implemented procedure is based on the selection of suitable sparsity-promoting TFR domains and on dedicated sample-point distribution strategies, achieving highly effective recovery results while starting from highly incomplete wave-field data.

Instructor: Prof. Damiana Lazzaro, University of Bologna (damiana.lazzaro@unibo.it)

Duration: 16 hours, 4 hour/week

Offered:
Wednesday, October 14, 10:00-13:00 (Aula Seminario I)
Friday, October 16, 10:00-13:00 (Aula Seminario II)
Monday, October 19, 10:00-12:00 (Aula Seminario II)
Wednesday, October 21, 10:00-13:00 (Aula Seminario I)
Friday, October 23, 10:00-13:00 (Aula Seminario II)
Monday, October 26, 10:00-12:00 (Aula Seminario II)

Content:
1) Serie di Fourier
2) Trasformata di Fourier continua di una funzione 𝑓(𝑥)∈𝐿2(𝑅) e sue proprietà – Teoremi di convoluzione e Correlazione
3) Trasformata Discreta di Fourier e sue proprietà – Teorema di Campionamento
4) Algoritmo Fast Fourier Transform (FFT)
5) Trasformata di Fourier continua di una funzione 𝑓(𝑥)∈𝐿2(𝑅2)
6) Trasformata di Fourier discreta 2D
7) Utilizzo dell’analisi di Fourier per l’enhancement di Immagini
8) Pregi e limiti dell’analisi di Fourier
9) Trasformata di Gabor ( Short Time Fourier Transform) – Principio di indeterminazione di Heisenberg
10) Funzioni Wavelet
11) Multiresolution Analysis
12) Algoritmi piramidali di Analisi e Sintesi
13) Utilizzo dell’analisi wavelet nell’ Image Processing
Sperimentazione in Matlab delle tecniche spiegate durante le lezioni frontali.

Instructor: Prof. Alessandro Marzani, University of Bologna (alessandro.marzani@unibo.it) and Dr. Antonio Palermo,, University of Bologna (antonio.palermo6@unibo.it)

Duration: 4 hours

Offered: June 2020

Abstract:
Many natural and engineered materials across different length scales present a geometry or a material composition with a periodic pattern. Originally, periodicity in both structural systems and engineered materials was introduced to achieve enhanced mechanical properties at a reduced weight. More recently periodicity has been utilized to design materials, structures or devices able to control the propagation of mechanical vibrations and waves. Periodic structures present, in fact, a unique dynamic response that can be exploited to impede the propagation of waves within desired frequency ranges and/or spatial directions. This course aims at introducing the fundamental aspects of elastic wave propagation in periodic structures. Basic concepts of lattice dynamics and wave propagation will be introduced by describing the response of 1D and 2D discrete systems. Dispersive features characteristic of periodic systems like bandgaps and wave beaming will be discussed. Additionally, appropriate computational techniques to extract the band structures properties of continuos periodic systems will be presented. In particular, the course will focus on Finite Element based techniques for band structure calculation, as the Wave Finite Element method and the Bloch operator Finite Element method.

Short course by Prof. Luca De Marchi, University of Bologna, offered on January 2020

further info: https://phd.unibo.it/eit4se[…]essing-for-guided-waves-shm

MATLAB course by Dr. Marinella Masina, University of Bologna, offered on February/March 2018

further info: https://phd.unibo.it/[…]/matlab-course-for-phd-students