Óscar Déniz

From spectral indices to actionable insights: sensitivity analysis of a multispectral U-Net for spatially-optimized urban heat island mitigation

Óscar Déniz, Hüseyin Pekkan, Ahmet Murat Özbayoğlu

Urban heat island (UHI) effects present significant challenges for sustainable urban development. Within the UDENE project, a deep-learning framework was established to model land surface temperature (LST) from remotely sensed urban indices, assess the effectiveness of green and blue infrastructure interventions such …

tr (code pays fourni par la source)

Deep learning framework for urban seismic risk assessment: a two-phase similarity algorithm for damage prediction and loss estimation across European-Mediterranean cities

Óscar Déniz, Hüseyin Pekkan, Ahmet Murat Özbayoğlu, Onur Lenk et autres

We present an integrated, two-step framework for high-resolution urban seismic risk assessment and similarity-based damage prediction developed under the UDENE initiative. The methodology couples an Earthquake Hazard Assessment, producing rasterized peak ground acceleration, spectral acceleration and intensity fields at 150 arcsec grid …

Spatio-temporal graph neural networks for human–AI collaborative decision-making

Israel Mateos-Aparicio-Ruiz, Pedro Montealegre-Macias, Óscar Déniz, Gloria Bueno

Collaborative decision-making (CDM) is essential in different domains where integrating diverse perspectives improves classification accuracy. Traditional aggregation methods, such as majority voting (MV), are static and fail to capture the dynamic, real-time interactions among decision-makers. We propose a task- and label-independent framework …

es (code pays fourni par la source)

Supplementary Material- MFusionJ: Multi-Focus and Multi-exposure Image Fusion, Enhancing Detail Preservation

Harbinder Singh, Jesús Ruiz-Santaquiteria, Gabriel Cristóbal, Kamalpreet Singh et autres

Multi-focus image fusion (MFIF) and multi-exposure image fusion (MEIF) results were generated using MFusionJ, an ImageJ plugin designed to provide a robust solution for depth-of-field (DoF) extension through advanced image processing. The plugin was also applied to multi-exposure image stacks, showcasing its …

A Novel Heuristic Algorithm for Scalable and High-Performance Automated Single-Line Diagram Generation in Electrical Grids

Óscar Déniz, Süha Nur Arslan

Automated generation of single line diagrams (SLDs) from power distribution network Geographic Information System (GIS) data remains a computational challenge requiring efficient algorithms that can handle networks of varying complexity. A novel smart bottom-up heuristic approach is developed to address these challenges …

de (code pays fourni par la source)

Simultaneous Robustness and Generalization Using Nearest Neighbor Classifiers

Óscar Déniz, Gloria Bueno, Aníbal Pedraza, Harbinder Singh

es (code pays fourni par la source)

Enhancing Collaborative Image Classification via Spatio-Temporal Graph Neural Networks: A Proof-of-concept Study on Human Group Decisions

Israel Mateos-Aparicio-Ruiz, Pedro Montealegre-Macias, Óscar Déniz, Aníbal Pedraza et autres

es (code pays fourni par la source)

Large Language Model Based Data Augmentation for Turkish Named Entity Recognition

Óscar Déniz, N. Ceyda Ünsoy, Bahaeddin Eravcı

Named Entity Recognition (NER) plays a fundamental role in identifying and classifying named entities within texts. However, in resource-scarce languages and applications—particularly in Turkish—the lack of annotated data leads to a decline in model performance. In this study, synthetic examples were generated …

tr (code pays fourni par la source)

Detection of adversarial examples through chaos quantification in time-series analysis

Harbinder Singh, Aníbal Pedraza, Óscar Déniz, Gloria Bueno

Abstract In the realm of deep learning, deep neural networks (DNNs) have recently propelled significant advancements in image classification applications. However, these DNN models are vulnerable to adversarial examples (AE), which are crafted by introducing imperceptible perturbations to legitimate samples, leading the …

es (code pays fourni par la source)

DeepGun: Deep Feature-Driven One-Class Classifier for Firearm Detection Using Visual Gun Features and Human Body Pose Estimation

Harbinder Singh, Óscar Déniz, Jesús Ruiz-Santaquiteria, Juan D. Muñoz et autres

The increasing frequency of mass shootings at public events and public buildings underscores the limitations of traditional surveillance systems, which rely on human operators monitoring multiple screens. Delayed response times often hinder security teams from intervening before an attack unfolds. Since firearms …

es (code pays fourni par la source)

Comparative analysis of deep learning methods for breast ultrasound lesion detection and classification

Noelia Vállez, Israel Mateos-Aparicio-Ruiz, Miguel Ángel Rienda, Óscar Déniz et autres

es (code pays fourni par la source)

Adversarial Examples Detection with Chaos-Based Multivariate Features

Harbinder Singh, Aníbal Pedraza, Óscar Déniz, Gloria Bueno

Abstract Deep neural networks (DNNs) have demonstrated strong performance in classification-based applications in the field of machine learning (ML). A DNN model is nonetheless susceptible to adversarial examples (AE), which are created by introducing minor well-designed changes to a regular example. In …

es (code pays fourni par la source)