Automatic detection of the inferior alveolar nerve in CT scans with CenterNet

The inferior alveolar nerve (IAN) lies within the mandibular canal, named inferior alveolar canal in literature. The detection of this nerve is important during maxillofacial surgeries or for creating dental implants. The poor quality of cone-beam computed tomography (CBCT) and computed tomography (CT) scans and/or bone gaps within the mandible increase the difficulty of this task, posing a challenge to human experts who are going to manually detect it and resulting in a time-consuming task.Therefore this thesis investigates two methods to automatically detect the IAN: a non-data driven technique and a deep-learning method. The latter tracks the IAN position at each frame leveraging detections obtained with the deep neural network CenterNet, fined-tuned for our task, and temporal and spatial information.

Ict security: Defence strategies against targeted attacks

L’avanzare delle tecnologie ICT e l’abbattimento dei costi di produzione hanno portato ad un aumento notevole della criminalità informatica. Tuttavia il cambiamento non è stato solamente quantitativo, infatti si può assistere ad un paradigm-shift degli attacchi informatici da completamente opportunistici, ovvero senza un target specifico, ad attacchi mirati aventi come obiettivo una particolare persona, impresa o nazione. Lo scopo della mia tesi è quello di analizzare modelli e tassonomie sia di attacco che di difesa, per poi valutare una effettiva strategia di difesa contro gli attacchi mirati. Il lavoro è stato svolto in un contesto aziendale come parte di un tirocinio. Come incipit, ho effettuato un attacco mirato contro l’azienda in questione per valutare la validità dei sistemi di difesa. L’attacco ha avuto successo, dimostrando l’inefficacia di moderni sistemi di difesa. Analizzando i motivi del fallimento nel rilevare l’attacco, sono giunto a formulare una strategia di difesa contro attacchi mirati sotto forma di servizio piuttosto che di prodotto. La mia proposta è un framework concettuale, chiamato WASTE (Warning Automatic System for Targeted Events) il cui scopo è fornire warnings ad un team di analisti a partire da eventi non sospetti, ed un business process che ho nominato HAZARD (Hacking Approach for Zealot Attack Response and Detection), che modella il servizio completo di difesa contro i targeted attack. Infine ho applicato il processo all’interno dell’azienda per mitigare minacce ed attacchi informatici.

UWB radar sensor networks: Detection algorithms design and experimental analysis

In the last years radar sensor networks for localization and tracking in indoor environment have generated more and more interest, especially for anti-intrusion security systems. These networks often use Ultra Wide Band (UWB) technology, which consists in sending very short (few nanoseconds) impulse signals. This approach guarantees high resolution and accuracy and also other advantages such as low price, low power consumption and narrow-band interference (jamming) robustness. In this thesis the overall data processing (done in MATLAB environment) is discussed, starting from experimental measures from sensor devices, ending with the 2D visualization of targets movements over time and focusing mainly on detection and localization algorithms. Moreover, two different scenarios and both single and multiple target tracking are analyzed.

Deep learning models for the automatic analysis of volleyball plays

Collecting and analysing data is an important element in any field of human activity and research. Even in sports, collecting and analyzing statistical data is attracting a growing interest. Some exemplar use cases are: improvement of technical/tactical aspects for team coaches, definition of game strategies based on the opposite team play or evaluation of the performance of players. Other advantages are related to taking more precise and impartial judgment in referee decisions: a wrong decision can change the outcomes of important matches. Finally, it can be useful to provide better representations and graphic effects that make the game more engaging for the audience during the match. Nowadays it is possible to delegate this type of task to automatic software systems that can use cameras or even hardware sensors to collect images or data and process them. One of the most efficient methods to collect data is to process the video images of the sporting event through mixed techniques concerning machine learning applied to computer vision. As in other domains in which computer vision can be applied, the main tasks in sports are related to object detection, player tracking, and to the pose estimation of athletes. The goal of the present thesis is to apply different models of CNNs to analyze volleyball matches. Starting from video frames of a volleyball match, we reproduce a bird's eye view of the playing court where all the players are projected, reporting also for each player the type of action she/he is performing.

Production Lot Detection through Synthetic Generated Data

Correctness of information gathered in production environments is an essential part of quality assurance processes in many industries, this task is often performed by human resources who visually take annotations in various steps of the production flow. Depending on the performed task the correlation between where exactly the information is gathered and what it represents is more than often lost in the process. The lack of labeled data places a great boundary on the application of deep neural networks aimed at object detection tasks, moreover supervised training of deep models requires a great amount of data to be available. Reaching an adequate large collection of labeled images through classic techniques of data annotations is an exhausting and costly task to perform, not always suitable for every scenario. A possible solution is to generate synthetic data that replicates the real one and use it to fine-tune a deep neural network trained on one or more source domains to a different target domain. The purpose of this thesis is to show a real case scenario where the provided data were both in great scarcity and missing the required annotations. Sequentially a possible approach is presented where synthetic data has been generated to address those issues while standing as a training base of deep neural networks for object detection, capable of working on images taken in production-like environments. Lastly, it compares performance on different types of synthetic data and convolutional neural networks used as backbones for the model.