A5011575105- Biometric Identification and Security
- Forensic Fingerprint Detection Methods
- Digital Media Forensic Detection
- Forensic and Genetic Research
- Face recognition and analysis
- EEG and Brain-Computer Interfaces
- Sleep and Work-Related Fatigue
- User Authentication and Security Systems
- Anomaly Detection Techniques and Applications
- Advanced Steganography and Watermarking Techniques
- Generative Adversarial Networks and Image Synthesis
- Face and Expression Recognition
- Gaze Tracking and Assistive Technology
- Speech Recognition and Synthesis
- Advanced Malware Detection Techniques
- Digital and Cyber Forensics
- Hate Speech and Cyberbullying Detection
- AI in cancer detection
- Emotion and Mood Recognition
- Video Surveillance and Tracking Methods
- Bullying, Victimization, and Aggression
- Advanced Image and Video Retrieval Techniques
- Context-Aware Activity Recognition Systems
University of Cagliari
2018-2025
The International Fingerprint liveness Detection Competition (LivDet) is an open and well-acknowledged meeting point of academies private companies that deal with the problem distinguishing images coming from reproductions fingerprints made artificial materials relative to real fingerprints. In this edition LivDet we invited competitors propose integrated algorithms matching systems. goal was investigate at which extent integration impact on whole performance. Twelve were submitted...
The harmful utilization of DeepFake technology poses a significant threat to public welfare, precipitating crisis in opinion. Existing detection methodologies, predominantly relying on convolutional neural networks and deep learning paradigms, focus achieving high in-domain recognition accuracy amidst many forgery techniques. However, overseeing the intricate interplay between textures artifacts results compromised performance across diverse scenarios. This paper introduces groundbreaking...
The rise of AI-generated synthetic media, or deepfakes, has introduced unprecedented opportunities and challenges across various fields, including entertainment, cybersecurity, digital communication. Using advanced frameworks such as Generative Adversarial Networks (GANs) Diffusion Models (DMs), deepfakes are capable producing highly realistic yet fabricated content, while these advancements enable creative innovative applications, they also pose severe ethical, social, security risks due to...
The International Fingerprint Liveness Detection Competition is an international biennial competition open to academia and industry with the aim assess report advances in Presentation Attack Detection. proposed "Liveness Action" "Fingerprint representation" challenges were aimed evaluate impact of a PAD embedded into verification system, effectiveness compactness feature sets for mobile applications. Furthermore, we experimented new spoof fabrication method that has particularly affected...
Bullying and cyberbullying are harmful social phenomena that involve the intentional, repeated use of power to intimidate or harm others. The ramifications these actions felt not just at individual level but also pervasively throughout society, necessitating immediate attention practical solutions. BullyBuster project pioneers a multi-disciplinary approach, integrating artificial intelligence (AI) techniques with psychological models comprehensively understand combat issues. In particular,...
AI-generated synthetic media, also called Deepfakes, have significantly influenced so many domains, from entertainment to cybersecurity. Generative Adversarial Networks (GANs) and Diffusion Models (DMs) are the main frameworks used create producing highly realistic yet fabricated content. While these technologies open up new creative possibilities, they bring substantial ethical security risks due their potential misuse. The rise of such advanced media has led development a cognitive bias...
The International Fingerprint Liveness Detection Competition (LivDet) is a biennial event that invites academic and industry participants to prove their advancements in Presentation Attack (PAD). This edition, LivDet2023, proposed two challenges, "Liveness Action" "Fingerprint Representation", evaluate the efficacy of PAD embedded verification systems effectiveness compactness feature sets. A third, "hidden" challenge inclusion subsets training set whose sensor information unknown, testing...
The usual trend for the conventional palmvein recognition techniques is first to extract discriminative hand-crafted feature representations from raw images, and then feed a classifier with them. Unfortunately, it not yet clear how effectiveness of such features may be held in case large user population or environments where variability among acquisitions same person increase. In order face this problem, considered that use multiple classifiers increase performance respect best individual...
The diffusion of fingerprint verification systems for security applications makes it urgent to investigate the embedding software-based presentation attack detection algorithms (PAD) into such systems. Companies and institutions need know whether integration would make system more "secure" technology available is ready, and, if so, at what operational working conditions. Despite significant improvements, especially by adopting deep learning approaches PAD, current research did not state much...
The assessment of the fingerprint PADs embedded into a comparison system represents an emerging topic in biometric recognition. Providing models and methods for this aim helps scientists, technologists, companies to simulate multiple scenarios have realistic view process's consequences on recognition system. most recent aimed at deriving overall performance, especially sequential liveness pointed out significant decrease Genuine Acceptance Rate (GAR). In particular, our previous studies...
The impact of voice disorders is becoming more widely acknowledged as a public health issue. Several machine learning-based classifiers with the potential to identify have been used in recent studies differentiate between normal and pathological voices sounds. In this paper, we focus on analyzing vulnerabilities these systems by exploring possibility attacks that can reverse classification compromise their reliability. Given critical nature personal information, understanding which types are...
This study examines the privacy implications of using Human Activity Recognition (HAR) features for violent action detection in fight against bullying. Given sensitive nature such applications, these systems must comply with regulations, as General Data Protection Regulation (GDPR). Our analysis focuses on whether commonly employed HAR particularly regarding risk inadvertently exposing personal identities when data includes biometric information. The findings, derived from development a...
The International Fingerprint Liveness Detection Competition (LivDet) is a biennial event that invites academic and industry participants to prove their advancements in Presentation Attack (PAD). This edition, LivDet2023, proposed two challenges, Action Representation, evaluate the efficacy of PAD embedded verification systems effectiveness compactness feature sets. A third, hidden challenge inclusion subsets training set whose sensor information unknown, testing ability generalize models....
In this study we investigate a textural processing method of electroencephalography (EEG) signal as an indicator to estimate the driver's vigilance in hypothetical Brain-Computer Interface (BCI) system. The novelty solution proposed relies on employing one-dimensional Local Binary Pattern (1D-LBP) algorithm for feature extraction from pre-processed EEG data. From resulting vector, classification is done according three classes: awake, tired and drowsy. claim that class transitions can be...
In the last five years, deep learning methods, in particular CNN, have attracted considerable attention field of face-based recognition, achieving impressive results. Despite this progress, it is not yet clear precisely to what extent features are able follow all intra-class variations that face can present over time. paper we investigate performance improvement recognition systems by adopting self updating strategies templates. For purpose, evaluate a well-known deep-learning...
Fingerprint authentication systems are highly vulnerable to artificial reproductions of fingerprint, called fingerprint presentation attacks. Detecting attacks is not trivial because attackers refine their replication techniques from year year. The International liveness Detection Competition (LivDet), an open and well-acknowledged meeting point academies private companies that deal with the problem attack detection, has goal assess performance detection (FPAD) algorithms by using standard...
The problem of interoperability is still open in fingerprint presentation attack detection (PAD) systems. This involves costs for designers and manufacturers who intend to change sensors personal recognition systems or design multi-sensor systems, because they need obtain sensor-specific spoofs retrain the system. solutions proposed state art mitigate require data from target sensor are therefore not exempt obtaining new data. In this paper, we provide insights PAD thanks an overview...
In this study we investigate a textural processing method of electroencephalography (EEG) signal as an indicator to estimate the driver's vigilance in hypothetical Brain-Computer Interface (BCI) system. The novelty solution proposed relies on employing one-dimensional Local Binary Pattern (1D-LBP) algorithm for feature extraction from pre-processed EEG data. From resulting vector, classification is done according three classes: awake, tired and drowsy. claim that class transitions can be...