arXiv subjects

Jean-Christophe Nebel

Publications and source records attributed to Jean-Christophe Nebel.

8 recordsLinked to original sources

arXiv2025

Language modelling techniques for analysing the impact of human genetic variation

Interpreting the effects of variants within the human genome and proteome is essential for analysing disease risk, predicting medication response, and developing personalised health interventions. Due to the intrinsic similarities between the structure of natural languages and genetic sequences, natural language processing techniques have demonstrated great applicability in computational variant effect prediction. In particular, the advent of the Transformer has led to significant advancements in the field. However, Transformer-based models are not without their limitations, and a number of extensions and alternatives have been developed to improve results and enhance computational efficiency. This review explores the use of language models for computational variant effect prediction over the past decade, analysing the main architectures, and identifying key trends and future directions.

cs.CL↗

arXiv2024

Steganography and Probabilistic Risk Analysis: A Game Theoretical Framework for Quantifying Adversary Advantage and Impact

In environments where adversaries engage in active surveillance and covert communication, defenders face the dual challenge of when to deploy steganography and whether it yields operational benefit. We present a novel game-theoretic model of steganographic operations that captures strategic interactions between a defender and an adversary through calibrated monetary primitives and nonlinear utility mappings. We derive mixed-strategy equilibria that drive conditional and unconditional success rates for hiding and detection, and introduce a time-varying adversarial advantage metric that quantifies when an attacker's incentives temporarily exceed the defender's detection capacity. By linking this advantage to a new currency-unit risk measure, we extend the classical risk formula into a decision-aware, monetised framework. A Monte Carlo simulation pipeline embeds payload shifts, detector learning, and scenario uncertainty to deliver distributions of success probabilities and expected losses rather than a static snapshot. Empirical calibration using breach cost statistics, regulatory fine caps, and expert elicitations supports strategic prescriptions for steganographic deployment, detector investment, and governance trade-offs. Our results provide actionable insight into when steganography strengthens organisational resilience, and when it may yield marginal or negative value.

cs.GT↗

arXiv2024

Evaluation of Environmental Conditions on Object Detection using Oriented Bounding Boxes for AR Applications

The objective of augmented reality (AR) is to add digital content to natural images and videos to create an interactive experience between the user and the environment. Scene analysis and object recognition play a crucial role in AR, as they must be performed quickly and accurately. In this study, a new approach is proposed that involves using oriented bounding boxes with a detection and recognition deep network to improve performance and processing time. The approach is evaluated using two datasets: a real image dataset (DOTA dataset) commonly used for computer vision tasks, and a synthetic dataset that simulates different environmental, lighting, and acquisition conditions. The focus of the evaluation is on small objects, which are difficult to detect and recognise. The results indicate that the proposed approach tends to produce better Average Precision and greater accuracy for small objects in most of the tested conditions.

cs.CV↗

arXiv2024

Visible and infrared self-supervised fusion trained on a single example

Multispectral imaging is an important task of image processing and computer vision, which is especially relevant to applications such as dehazing or object detection. With the development of the RGBT (RGB & Thermal) sensor, the problem of visible (RGB) to Near Infrared (NIR) image fusion has become particularly timely. Indeed, while visible images see color, but suffer from noise, haze, and clouds, the NIR channel captures a clearer picture. The proposed approach fuses these two channels by training a Convolutional Neural Network by Self Supervised Learning (SSL) on a single example. For each such pair, RGB and NIR, the network is trained for seconds to deduce the final fusion. The SSL is based on the comparison of the Structure of Similarity and Edge-Preservation losses, where the labels for the SSL are the input channels themselves. This fusion preserves the relevant detail of each spectral channel without relying on a heavy training process. Experiments demonstrate that the proposed approach achieves similar or better qualitative and quantitative multispectral fusion results than other state-of-the-art methods that do not rely on heavy training and/or large datasets.

cs.CV↗

arXiv2023

Multispectral image fusion based on super pixel segmentation

Multispectral image fusion is a computer vision process that is essential to remote sensing. For applications such as dehazing and object detection, there is a need to offer solutions that can perform in real-time on any type of scene. Unfortunately, current state-of-the-art approaches do not meet these criteria as they need to be trained on domain-specific data and have high computational complexity. This paper focuses on the task of fusing color (RGB) and near-infrared (NIR) images as this the typical RGBT sensors, as in multispectral cameras for detection, fusion, and dehazing. Indeed, the NIR channel has the ability to capture details not visible in RGB and see beyond haze, fog, and clouds. To combine this information, a novel approach based on superpixel segmentation is designed so that multispectral image fusion is performed according to the specific local content of the images to be fused. Therefore, the proposed method produces a fusion that contains the most relevant content of each spectrum. The experiments reported in this manuscript show that the novel approach better preserve details than alternative fusion methods.

cs.CV↗

arXiv2021

Classic versus deep learning approaches to address computer vision challenges

Computer vision and image processing address many challenging applications. While the last decade has seen deep neural network architectures revolutionizing those fields, early methods relied on 'classic', i.e., non-learned approaches. In this study, we explore the differences between classic and deep learning (DL) algorithms to gain new insight regarding which is more suitable for a given application. The focus is on two challenging ill-posed problems, namely faint edge detection and multispectral image registration, studying recent state-of-the-art DL and classic solutions. While those DL algorithms outperform classic methods in terms of accuracy and development time, they tend to have higher resource requirements and are unable to perform outside their training space. Moreover, classic algorithms are more transparent, which facilitates their adoption for real-life applications. As both classes of approaches have unique strengths and limitations, the choice of a solution is clearly application dependent.

cs.CV↗

arXiv2021

Game theory to enhance stock management of Personal Protective Equipment (PPE) during the COVID-19 outbreak

Since the outbreak of the COVID-19 pandemic, many healthcare facilities have suffered from shortages in medical resources, particularly in Personal Protective Equipment (PPE). In this paper, we propose a game-theoretic approach to schedule PPE orders among healthcare facilities. In this PPE game, each independent healthcare facility optimises its own storage utilisation in order to keep its PPE cost at a minimum. Such a model can reduce peak demand considerably when applied to a variable PPE consumption profile. Experiments conducted for NHS England regions using actual data confirm that the challenge of securing PPE supply during disasters such as COVID-19 can be eased if proper stock management procedures are adopted. These procedures can include early stockpiling, increasing storage capacities and implementing measures that can prolong the time period between successive infection waves, such as social distancing measures. Simulation results suggest that the provision of PPE dedicated storage space can be a viable solution to avoid straining PPE supply chains in case a second wave of COVID-19 infections occurs.

cs.CY↗

arXiv2018

Security Attacks on Smart Grid Scheduling and Their Defences: A Game-Theoretic Approach

The introduction of advanced communication infrastructure into the power grid raises a plethora of new opportunities to tackle climate change. This paper is concerned with the security of energy management systems which are expected to be implemented in the future smart grid. The existence of a novel class of false data injection attacks that are based on modifying forecasted demand data is demonstrated, and the impact of the attacks on a typical system's parameters is identified, using a simulated scenario. Monitoring strategies that the utility company may employ in order to detect the attacks are proposed and a game--theoretic approach is used to support the utility company's decision--making process for the allocation of their defence resources. Informed by these findings, a generic security game is devised and solved, revealing the existence of several Nash Equilibrium strategies. The practical outcomes of these results for the utility company are discussed in detail and a proposal is made, suggesting how the generic model may be applied to other scenarios.

cs.GT↗