Abstract
Face recognition technology, despite its advancements and increasing accuracy, still presents significant challenges in explainability and ethical concerns, especially when applied in sensitive domains such as surveillance, law enforcement, and access control. The opaque nature of deep learning models jeopardises transparency, bias, and user trust. Concurrently, the proliferation of web applications presents a unique opportunity to develop accessible and interactive tools for demonstrating and analysing these complex systems. These tools can facilitate model decision exploration with various images, aiding in bias mitigation or enhancing users’ trust by allowing them to see the model in action and understand its reasoning. We propose an explainable face recognition web application designed to support enrolment, identification, authentication, and verification while providing visual explanations through pixel-wise importance maps to clarify the model’s decision-making process. The system is built in compliance with the European Union General Data Protection Regulation, ensuring data privacy and user control over personal information. The application is also designed for scalability, capable of efficiently managing large datasets. Load tests conducted on databases containing up to 1,000,000 images confirm its efficiency. This scalability ensures robust performance and a seamless user experience even with database growth. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

Abstract
The phase-matching conditions for exciting surface plasmon resonances (SPR) in plasmonic films are typically satisfied via prism, optical fibers or grating-assisted coupling. We recently showed that plasmonic nanospheres can act as local emitters, exciting SPR waves on thin films-termed nanoparticle-induced SPR (NPI-SPR). This structure holds promise for sensing, but the effects of optical fiber geometry and nanoparticle anisotropy on the response were unexplored. This study examines these factors, showing that an etched multimode fiber with a 200 mu m core diameter, taper ratio of 4, and etching angle of 20 degrees optimizes interaction with plasmonic nanoparticles. Tuning the nanoparticle aspect ratio from 1 to 3 shifts the NPI-SPR band from 780 to 1580 nm, with excitation highly dependent on the incident light angle. Notably, for light incident parallel to the film plane, a refractive index sensitivity exceeding 1000 nm/RIU is achieved. This efficient light emission combines the field locality enhancements of plasmonic nanoparticle-on-film structures with the emission efficiency of plasmonic nanoantennas, advancing plasmonic optical fiber chemical and biosensors.

Abstract

Abstract
Optical Wireless Communications (OWC) has recently emerged as a viable alternative to radio-frequency technology, especially for the Internet of Things (IoT) domain. However, current simulation tools primarily focus on physical layer modelling, ignoring network-level issues and energy-constrained environments. This paper presents an energy-aware OWC module for ns-3 that addresses these limitations. The module includes specific PHY and MAC layers and integrates an energy model, a mobility model, and models of monochromatic transceivers and photodetectors, supporting both visible light and infrared (IR) communications. Verification against MATLAB simulations confirms the accuracy of our implementation. Additionally, mobility tests demonstrate that an energy-restricted end device transmitting via IR can maintain a stable connection with a gateway at distances up to 2.5 m, provided the SNR is above 10 dB. These results confirm the capabilities of our module and its potential to facilitate the development of energy-efficient OWC-based IoT systems.

Abstract
Reinforced concrete structures form the backbone of civil infrastructure due to their durability, longevity, affordability, and availability. However, aging concrete poses challenges, with decay often beginning internally and becoming visible only at advanced stages, leading to costly repairs, restricted functionality, and safety risks. To address these challenges, sensors are crucial for enhancing infrastructure resilience and optimizing repairs. This study employs multimode optical fibers to monitor concrete curing, water ingress, relative humidity (RH), cement paste carbonation, and rebar corrosion. Four sensors monitor changes in reflection at the fiber tip of a 600 mu m multimode fiber (MMF) using LEDs and photodiodes, connected via a fiber bundle containing two 200 mu m MMF. Variations in the refractive index around the fiber tip are used to monitor water throughout the concrete lifecycle, including curing, RH changes and water intrusion. Colorimetric changes in a cement paste layer and an iron-thin film are used to monitor carbonation and corrosion. The curing sensor is temperature-independent and correlates strongly with cumulative heat release from cement hydration (R=0.95). The RH sensor monitors up to and beyond 100% RH, detecting water intrusion. The corrosion sensor detects early corrosion stages and distinguishes between reflection losses from corrosion and mechanical changes. The layer of cement paste for carbonation monitoring increases reflected intensity by 17% due to carbonation, with 63% of the increase occurring in 80 minutes in a 3% CO2 atmosphere. The broad monitoring scope and low implementation cost make this sensor a unique option among commercially available solutions for structural health monitoring of reinforced concrete.

Abstract