17628 research outputs found
Sort by
A Comprehensive Framework for Out-of-Distribution Detection and Open-Set Recognition in SAR Targets
Get PDFThe rejection of outlier data in synthetic aperture radar (SAR) image analysis presents a significant challenge, particularly in the scenarios of out-of-distribution (OOD) detection and open set recognition (OSR). This issue arises due to the constant emergence of new categories in the real world and the fact that available datasets often do not provide comprehensive coverage of these new categories, resulting in models that lack the ability to effectively recognize and adapt in the face of outlier data. Existing methods struggle to establish clear decision boundaries between categories, primarily because they lack the capacity to capture the complex and diverse feature distributions inherent in SAR data. In addition, insufficient modeling of neuron activations affects the accurate discrimination between in-distribution (InD) and OOD data. To address these issues, we propose a comprehensive framework for OOD detection / OSR of SAR targets based on neuron coverage and outlier activation analysis, which refines the representation of outlier categories and sharpens the decision boundaries, ensuring a more precise demarcation between InD and OOD categories. First, we define the neuron activation states by considering the outputs of neurons and their influence on the model’s decision-making process. The varying activation coverage reflects how the network responds to different types of inputs within its parameter space. Next, we design a training method that simulates OOD data. By generating low-probability density points near decision boundaries using a multivariate normal distribution and perturbing them with noise. Finally, a regularization term based on nearest-neighbor distances is introduced to refine the scores of both InD and OOD data, thereby facilitating the effective rejection of outlier categories. Experimental results on multiple SAR datasets demonstrate that our approach significantly outperforms existing methods in key performance metrics, offering a more effective solution for outlier category rejection in SAR image analysis
SAD-GAN: A Novel Secure Anomaly Detection Framework for Enhancing the Resilience of Cyber-Physical Systems
No full textCyber-physical systems occupy a significant portion of the critical infrastructure market, but their prominence has raised concerns due to their susceptibility to certain anomalies. The typical approaches tend to be ineffective to flexible and complex conditions of CPS environments. To address these issues, this paper presents SAD-GAN—self-adaptive deep generative adversarial network—framework that aimed at improving real-time detection of anomalies. SAD-GAN follows a GAN framework with generator (G), which is trained to generate normal behavior of the system, and discriminator (D) which is trained to distinguish normal and artificial data patterns. The anomalies are detected based on a dual-scoring mechanism which consists of reconstruction error and discriminator confidence and are multiplied by the two adjustable constants, 2 and 3. These coefficients determine the relative adjustment of action of each of the scores of the final anomaly detection process and are pumped dynamically with the verification turnover to guarantee credible detection with the changes in the progress of the system. This mechanism enables SAD-GAN to learn and adjust at run time with no need of manual reconfiguration. It was tested against benchmark CPS datasets (SWaT and WADI) and proved to be better performing than conventional models, e.g., Isolation Forest and static GANs. SAD-GAN has an accuracy of 97.2, and the false positive was under 2% and identified significant changes in the time of detection and flexibility. These findings validate the efficiency of SAD-GAN to find minute and changing anomalies without a high number of false alarms. The suggested method, in general, provides a flexible, smart, and adaptable algorithm of robust anomaly detection in contemporary CPS systems
Mortality analysis of patients with acute coronary syndrome receiving comprehensive cardiac care (KOS-Zawal) during the COVID-19 pandemic period
Get PDFIntroduction:Due to the SARS-CoV-2 pandemic, there have been fundamental changes to the delivery and operation of healthcare facilities across the world, significantly impacting how patients with a variety of diseases are treated. We aimed to assess the impact of the COVID-19 pandemic on patient management outcomes among patients with acute coronary syndromes (ACS) and explore the differences in patients who were treated within and outside the coordinated care programme for patients after ACS (KOS-Zawal).Material and methods:We analysed 472,996 medical records of patients after ACS from 2017 to 2022. The study examined information on deaths in two groups of patients: those included and those not included in the KOS-Zawal programme.Results:Before the COVID-19 pandemic a significantly higher mortality rate was observed in the group of patients not covered by the KOS-Zawal benefit compared with covered patients (25.5% vs. 15.8%; p < 0.0001). During the COVID-19 pandemic a significantly higher incidence of death was noted in the group of patients not covered by KOS-Zawal compared with patients covered by the programme (18% vs. 7.9%; p < 0.0001). Compared to the time before and during COVID-19, the number of deaths among patients not covered (25.5% vs. 18%; p < 0.0001) and covered by KOS-Zawal (15.8% vs. 7.9%, p < 0.0001) decreased significantly.Conclusions:Patients not covered by KOS-Zawal had a significantly higher mortality rate compared to those covered by the programme during the pandemic. The pandemic significantly affected patients under KOS-Zawal care, with a reduced mortality rate
Retiring for the Night: How Negative Attitudes Towards Ageing Can Shape Expectations and Attitudes Towards Sleep Among Adults Aged 60+
Get PDFIntroduction: Negative attitudes towards ageing are pervasive and can create significant barriers to quality of life and longevity for adults aged 60 and older. Studies indicate that older adults with negative views of ageing often perceive common, treatable ailments as a natural consequence of age, impacting their health-seeking behaviours, such as good sleep hygiene. Little is known about how these ageing attitudes shape beliefs about sleep in older adults. Methods: This study examined the association between attitudes towards ageing and beliefs about sleep in older adults, using data collected through online surveys. Measures included the Pittsburgh Sleep Quality Index, the Centre for Epidemiologic Studies Depression Scale, and the Brief Ageing Perceptions Questionnaire. Hierarchical regression analysis was applied to assess the predictive relationships among these variables. Results: Emotional responses to ageing, such as worry and frustration, were strongly associated with dysfunctional sleep beliefs, explaining additional variance in sleep beliefs even after controlling for demographic factors, depressive symptoms, and sleep quality. Control-related perceptions of ageing predicted locus of control for sleep, with positive beliefs associated with an internal locus of control and negative beliefs linked to an external locus of control. Conclusion: Findings suggest that perceptions of ageing significantly influence attitudes towards sleep, potentially affecting health-seeking behaviours. Addressing age-related stereotypes and fostering positive attitudes may reduce barriers to healthcare engagement and encourage healthier sleep behaviours in older adults, thereby supporting better mental and physical well-being. This study highlights the importance of tackling ageing stereotypes to promote healthier ageing
Social workers’ views and experiences of self-care practices: a qualitative interview study
Get PDFSelf-care is increasingly advocated as necessary for improving social workers’ wellbeing. However, it remains a contested term, with limited understanding of social workers’ views and experiences of what it constitutes in practice. This qualitative study employed semi-structured interviews with nine social workers from three local authorities in Scotland. Informed by vulnerability theory, a six-phase thematic analysis was applied to explore social workers’ views and experiences of self-care practices. Three key themes emerged: (1) understanding and conceptualizing self-care, illustrating practitioners’ perceptions of self-care as individualized, multifaceted strategies aimed at both personal wellbeing and professional efficacy, with heightened awareness since COVID-19; (2) the implementation paradox, highlighting fundamental tensions between acknowledging professional vulnerability and managing organizational demands, workload pressures, and insufficient institutional support; and (3) toward sustainable self-care practice, identifying pathways through deliberate individual practices, organizational support, educational preparation, and culturally-sensitive policies. Public health policymakers and healthcare organizations should prioritize structural reforms to enhance workforce resilience, thereby improving service quality, practitioner wellbeing, and overall public health outcomes
Data-Based Encryption Iterative Learning Heading Control for Unmanned Surface Vehicles
No full textThis letter investigates a data-driven iterative learning heading control problem for unmanned surface vehicles (USVs) with encoding-decoding mechanisms. First, a compact form dynamic linearized model of the USV is established using dynamic linearization techniques and redefined outputs. Then, an encoding-decoding scheme is designed, which encodes the data before transmission and decodes the data received by the controller. This strategy compresses the data and offers protection from potential breaches of information. Finally, the convergence of the designed method is theoretically analyzed, and simulation results demonstrate its effectiveness in enhancing heading control performance
Analysing the role of LLMs in cybersecurity incident management
No full textCybersecurity and artificial intelligence (AI) increasingly intersect as organizations grapple with sophisticated cyber threats and expanding digital landscapes. Incident response teams traditionally rely on structured procedures to identify, manage, and mitigate cyber incidents. Our work explores the effectiveness of generative AI, specifically Large Language Models (LLMs), within cybersecurity, focusing primarily on incident response processes. Experimental evaluations demonstrate that specific LLMs exhibit distinct strengths suitable for different stages of incident management. GPT-4o and GPT-3.5 show high clarity, consistency and coherence, making them appropriate for real-time containment, isolation, eradication and recovery tasks. Conversely, models such as GPT-o1 and GPT-4 offer superior reasoning capabilities and conciseness, better supporting incident preparation, post-incident analysis, vulnerability assessment and training development. Key limitations pertaining to current LLM implementations are identified, particularly token context constraints in addition to a discussion about ethical considerations regarding reliance on AI responses, including potential impacts on workforce skills and organizational security posture
Rumble in the jungle: Convolutional neural networks demonstrate accurate footfall identification of terrestrial mammals
Get PDF1. Accurate and resource-efficient wildlife survey methodologies are crucial to informing biodiversity, wildlife management and conservation planning. Large mammals are particularly challenging to survey due to their large spatial ranges and often-elusive behaviour. Recent exploration of new approaches has led to seismology being used primarily in elephant behavioural studies; however, little had been done to investigate whether this methodology could provide accurate identification of multiple species and of those with lower body masses.2. We further developed this seismological approach by utilising convolutional neural networks (CNN) to individually identify four mammal species from recordings of their footfall. To facilitate this work, we built a prototype footfall trap — a seismic node customised for recording footfall — which is both inexpensive and easily replicable. Footfall events were extracted using a fully automated workflow and used to train four CNN models.3. We extracted 10,965 footfall events from 9 days of recordings in a captive environment and observed vibrations detected within a 3m radius. The best-performing CNN model achieved a species identification accuracy of 92% with F1 scores for each species between 86 and 97. Disregarding low confidence predictions (<0.8) increased accuracy to 99% but resulted in 27% data loss. The prototype footfall trap costs approximately 220 USD; this is substantially cheaper than seismometers used in similar studies and could be further reduced.4. Solution. Our findings demonstrate that footfall traps have the potential to provide accurate species identification and could be fully automated once the necessary training data has been established. This methodology leverages some of the benefits of passive acoustic monitoring but for silent species, while overcoming some camera trap limitations in being omnidirectional and unrestricted by surrounding vegetation. Further research is required to understand effectiveness with a greater range of species, particularly taxonomically similar species, and how well the models generalise when facing other environmental conditions
Evaluation of the Mechanical and Physical Behaviors of Flax Fiber‐Reinforced Polybutylene Succinate Biodegradable Composites in Packaging Applications
Get PDFApplying surface modification routes to natural fibers is a practical option for solving the incompatibility problem in polymeric composites. This study aimed to improve the properties of biodegradable and environmentally friendly polybutylene succinate (PBS) composites involving flax fiber (FF) with unmodified, alkalization, and silanization surface modification applied at 20 wt%. The surface modifications were successfully characterized by FTIR and SEM. The composites were prepared using a twin‐screw extruder followed by injection molding. The mechanical, thermal, thermo‐mechanical, wear, water absorption, biodegradation, and morphological properties were evaluated. The incorporation of FF improved the mechanical and thermal performance of PBS, while the surface modifications further increased the fiber‐matrix adhesion. In particular, the silanized FF provided water resistance due to the hydrophobic nature of siloxane and exhibited the lowest biodegradation rate under fungal exposure. In the case of silanized FF, a 5.4% improvement in tensile strength and nearly 3% of water absorption capacity were reached. Based on these outcomes, the silanized FF‐filled PBS composite was suitable for packaging applications where biodegradable, water‐resistant, and mechanically strong materials are required
Data sources:
Access Repository Dashboard
Do you manage Repository@Napier? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!