A5101846914- Advanced Malware Detection Techniques
- Biometric Identification and Security
- Hand Gesture Recognition Systems
- User Authentication and Security Systems
- Human Pose and Action Recognition
- Cybercrime and Law Enforcement Studies
- Anomaly Detection Techniques and Applications
- Face recognition and analysis
- Digital Transformation in Industry
- Advanced X-ray and CT Imaging
- Crime, Illicit Activities, and Governance
- Adversarial Robustness in Machine Learning
- AI in cancer detection
- Virtual Reality Applications and Impacts
- Network Security and Intrusion Detection
- Artificial Intelligence in Healthcare
- Image Retrieval and Classification Techniques
- Crime Patterns and Interventions
- Diverse Approaches in Healthcare and Education Studies
- Advanced Steganography and Watermarking Techniques
- Parkinson's Disease Mechanisms and Treatments
- Complementary and Alternative Medicine Studies
- Neurological disorders and treatments
- Technology Assessment and Management
- Advanced Neural Network Applications
Harokopio University of Athens
2023-2024
Foundation for Research and Technology Hellas
2024
FORTH Institute of Computer Science
2024
Centre for Research and Technology Hellas
2019-2024
China Philanthropy Research Institute
2019-2023
Trilateral Research & Consulting
2023
University of Western Macedonia
2023
Information Technologies Institute
2023
Università Cattolica del Sacro Cuore
2023
Mediterranean University
2023
In this paper, a gaze-based Relevance Feedback (RF) approach to region-based image retrieval is presented. Fundamental idea of the proposed method comprises iterative estimation real-world objects (or their constituent parts) that are interest user and subsequent exploitation information for refining results. Primary novelties work are: a) introduction new set gaze features realizing user's relevance assessment prediction at region-level, b) design time-efficient effective object-based RF...
<p>Recent trends in the modus operandi of technologically-aware criminal groups engaged illicit goods trafficking (e.g., firearms, drugs, cultural artifacts, etc.) have given rise to significant security challenges. The use cryptocurrency-based payments, 3D printing, social media and/or Dark Web by organized crime leads transactions beyond reach authorities, thus opening up new business opportunities actors at expense greater societal good and rule law. As a result, lot scientific...
Automated visual firearms classification from RGB images is an important real-world task with applications in public space security, intelligence gathering and law enforcement investigations. When applied to massively crawled the World Wide Web (including social media dark sites), it can serve as component of systems that attempt identify criminal trafficking networks, by analyzing Big Data open-source intelligence. Deep Neural Networks (DNN) are state-of-the-art methodology for achieving...
Illicit object detection is a critical task performed at various high-security locations, including airports, train stations, subways, and ports. The continuous tedious work of examining thousands X-ray images per hour can be mentally taxing. Thus, Deep Neural Networks (DNNs) used to automate the image analysis process, improve efficiency alleviate security officers' inspection burden. neural architectures typically utilized in relevant literature are Convolutional (CNNs), with Vision...
The developing field of enhanced diagnostic techniques in the diagnosis infectious diseases, constitutes a crucial domain modern healthcare. By utilizing Gas Chromatography-Ion Mobility Spectrometry (GC-IMS) data and incorporating machine learning algorithms into one platform, our research aims to tackle ongoing issue precise infection identification. Inspired by these difficulties, goals consist creating strong analytics process, enhancing (ML) models, performing thorough validation for...
Self-Supervised Learning (SSL) is a valuable and robust training methodology for contemporary Deep Neural Networks (DNNs), enabling unsupervised pretraining on `pretext task' that does not require ground-truth labels/annotation. This allows efficient representation learning from massive amounts of unlabeled data, which in turn leads to increased accuracy `downstream by exploiting supervised transfer learning. Despite the relatively straightforward conceptualization applicability SSL, it...
Over the recent years, protection of so-called `soft-targets', i.e. locations easily accessible by general public with relatively low, though, security measures, has emerged as a rather challenging and increasingly important issue. The complexity seriousness this threat growths nowadays exponentially, due to emergence new advanced technologies (e.g. Artificial Intelligence (AI), Autonomous Vehicles (AVs), 3D printing, etc.); especially when it comes large-scale, popular diverse spaces. In...
<p>Recent trends in the modus operandi of technologically-aware criminal groups engaged illicit goods trafficking (e.g., firearms, drugs, cultural artifacts, etc.) have given rise to significant security challenges. The use cryptocurrency-based payments, 3D printing, social media and/or Dark Web by organized crime leads transactions beyond reach authorities, thus opening up new business opportunities actors at expense greater societal good and rule law. As a result, lot scientific...
Automated visual firearms classification from RGB images is an important real-world task with applications in public space security, intelligence gathering and law enforcement investigations. When applied to massively crawled the World Wide Web (including social media dark sites), it can serve as component of systems that attempt identify criminal trafficking networks, by analyzing Big Data open-source intelligence. Deep Neural Networks (DNN) are state-of-the-art methodology for achieving...
Automated detection of contraband items in X-ray images can significantly increase public safety, by enhancing the productivity and alleviating mental load security officers airports, subways, customs/post offices, etc. The large volume high throughput passengers, mailed parcels, etc., during rush hours practically make it a Big Data problem. Modern computer vision algorithms relying on Deep Neural Networks (DNNs) have proven capable undertaking this task even under resource-constrained...