Datos de investigación+información ...
Recursos didácticos+información ...
Resultados de investigación+información ...
Tesis+información ...
Trabajos fin de grado+información ...
Trabajos fin de máster+información ...
UNIR Editorial+información ...
UNIR Revistas+información ...
Publicaciones recientes
Acceso AbiertoItem type: Trabajo fin de Máster o TFM , 
Agregar a mi lista
Piloto Experimental del Algoritmo Xander para la Detección de Intrusiones en red utilizando la Computación Cuántica Híbrida(2026-02-11) Peñaloza-Araque, Óscar JavierEste Trabajo Fin de Máster presenta un piloto experimental para la detección de intrusiones en red mediante el algoritmo Xander, el cual representa una arquitectura híbrida cuántico-clásica, que integra modelos tales como; la Máquina de Vectores de Soporte Cuántico QSVM, el Circuito Cuántico Variacional VQC, y la Red Neuronal Convolucional Cuántica QCNN, en un ensamble con clasificadores tradicionales. El estudio utiliza el conjunto de datos CIC-IDS-2017 y desarrolla una metodología reproducible que comprende preprocesamiento, reducción de dimensionalidad y evaluación del sistema en tres niveles: Estimator local sin ruido, IBM Quantum Runtime Aer Simulator con aceleración clásica y ejecución en hardware cuántico real en entorno NISQ. Los resultados muestran métricas superiores al 99% en accuracy y F1-score, demostrando la viabilidad operativa del enfoque híbrido y la contribución complementaria del componente cuántico dentro del ensamble. Se concluye que Xander es una alternativa prometedora para mejorar la detección de intrusiones, especialmente en escenarios donde la robustez y la trazabilidad son críticas.- Acceso Abierto
Item type: Artículo , Agregar a mi lista
Demosaicking Algorithm Using Deep Residual Convolutional Network(IJIMAI, 2026-06-01) Jin Wang; Siyou Guo; Qilei Li; David Camacho; Gwanggil JeonSingle-sensor imaging systems are widely deployed in portable devices including digital cameras, smartphones, and personal digital assistants (PDAs) for real-time image acquisition. While convolutional neural networks (CNNs) have demonstrated exceptional capabilities in various image processing tasks, their potential for demosaicking applications remains underexplored. This paper presents a demosaicking framework utilizing a Deep Residual Convolutional Neural Network (DRCNN) architecture. Firstly, we initialize the mosaicked images using conventional demosaicking algorithms and learn the DRCNN for three color channels. The proposed DRCNN architecture innovatively integrates three core components: Binary Convolution Units (BCUs) for computational efficiency, Efficient Layer Aggregation Networks (ELAN) for multi-scale feature fusion, and Dense Residual Blocks (DRBs) for enhanced gradient flow. Comprehensive evaluations demonstrate that the proposed algorithms outperform existing approaches in PSNR, computational complexity, and visual quality. - Acceso Abierto
Item type: Artículo , Agregar a mi lista
Editor's Note(IJIMAI, 2026-06-01) Paulo Alonso Gaona-GarcíaRecent advances in Artificial Intelligence (AI) applied to Human–Computer Interaction (HCI) have been characterized by the convergence of immersive technologies, semantic reasoning, and data-centric learning paradigms. In augmented reality (AR), AI enables real-time interaction by integrating computer vision and contextual understanding into dynamic environments, although it simultaneously introduces new security vulnerabilities that require robust AI-driven protection mechanisms. In parallel, AI-driven cybersecurity has evolved toward proactive and adaptive models that leverage machine learning for real-time threat detection, anomaly analysis, and predictive mitigation, significantly enhancing the resilience of digital systems. Ontologies further strengthen this landscape by providing structured, semantically rich representations of cybersecurity knowledge, facilitating interoperability, automated reasoning, and intelligent threat assessment. Additionally, advances in visual analytics and time series foundation models are enabling the processing of large-scale temporal and multimodal datasets, supporting more sophisticated data interpretation and decision-making processes in complex environments. Modern computer vision systems leverage deep learning architectures such as convolutional neural networks and vision transformers to extract high-level features, enabling accurate object recognition, pattern detection, and scene understanding. Advances in multimodal AI systems—capable of integrating text, images, audio, and video—have become essential for achieving more comprehensive and context-aware intelligence, especially in domains such as medical imaging and interactive systems. Based on these advances highlight, this regular issue presents the most notable research in this direction.
ReUNIR se difunde en
.png)