Portail HAL de l'Université du Littoral Côte d'Opale
Not a member yet
11255 research outputs found
Sort by
Approche couplée VAE et interpolateur pour l'émulation rapide d'images hyperspectrales
No full textNational audienceLa technologie d'imagerie hyperspectrale joue un rôle crucial dans l'extraction d'informations précieuses du spectre électromagnétique, ce qui permet d'obtenir des cubes de données spatiales et spectrales détaillées appelées images hyperspectrales (HSI). Cette technologie trouve des applications dans divers domaines tels que l'astrophysique, la surveillance agricole et la télédétection. La génération de données hyperspectrales haute fidélité est un élément essentiel des missions d'observation, mais implique traditionnellement des simulations numériques coûteuses. Ce défi peut être relevé en construisant un modèle statistique qui se rapproche de la distribution des données simulées. Ce processus est appelé émulation. Cet article présente une nouvelle approche d'émulation hyperspectrale associant un autoencodeur variationnel (VAE) et un interpolateur neuronal pour l'émulation rapide de HSI. Le VAE est entraîné à reproduire des données, et un interpolateur est ensuite utilisé pour lier les paramètres biophysiques à l'espace latent appris par le VAE. Cette méthode permet l'échantillonnage de HSI à partir de paramètres biophysiques de manière efficace et précise. Cette approche est évaluée sur les images couleur de l’océan issues du satellite Sentinel-3 et montre des améliorations significatives en termes de précision, d’efficacité et de synthèse des informations au niveau de la scène par rapport aux méthodes existantes
On the Phase Transition of the Euclidean Travelling Salesman Problem with Time Windows
No full textInternational audienceAlgorithms are often evaluated on randomly generated instances to study scale-up properties with respect to features such as the size, for example. Also, machine learning based approaches often train models on randomly generated instances as they need large sets of training instances. In this paper, we consider the Euclidean Travelling Salesman Problem with Time Windows (TSPTW), and we study the impact of parameters used to randomly generate TSPTW instances on hardness and feasibility. We first consider the decision version of the problem, where feasibility depends on start and end times of time windows. We introduce two parameters, α and β, for controlling the tightness of the time horizon and the time windows. We show that instance hardness is related to a phase transition phenomenon: as we increase α and β, we pass from an unfeasible region (where almost all generated instances have no solution) to a feasible region (where almost all generated instances have solutions), and the hardest instances are located within the transition zone. We formally relate this transition zone with respect to α and β, thus allowing us to control hardness and feasibility when randomly generating instances. Then, we study the optimization problem, the goal of which is to find the smallest tour that satisfies all time windows. We show that the empirical hardness is still related to the phase transition: hardness increases when moving from the infeasible region to the transition zone, as in the decision problem. However, unlike the decision problem, some hard instances are also located in the feasible region where instances are very loosely constrained
Optimal Control of Medical Drug in a Nonlocal Model of Solid Tumor Growth
No full textInternational audienceThis paper presents a mathematical framework for optimizing drug delivery in cancer treatment using a nonlocal model of solid tumor growth. We present a coupled system of partial differential equations that incorporate long-range cellular interactions through integral terms and drug-induced cell death. The model accounts for spatial heterogeneity in both tumor cell density and drug concentration while capturing the complex dynamics of drug resistance development. We first establish the well-posedness of the coupled system by proving the existence and uniqueness of a solution under appropriate regularity conditions. The optimal control problem is then formulated to minimize tumor size while accounting for drug toxicity constraints. Using variational methods, we derive the necessary optimality conditions and characterize the optimal control through an adjoint system. Theoretical results can help to design effective chemotherapy schedules that balance treatment efficacy with adverse effects
Exploring the significance of different amendments to improve phytoremediation efficiency: focus on soil ecosystem services
No full textInternational audiencePhytoremediation is recognized as an environmentally, economically and socially efficient phytotechnology for the reclamation of polluted soils. To improve its efficiency, several strategies can be used including the optimization of agronomic practices, selection of high-performance plant species but also the application of amendments. Despite evidences of the benefits provided by different types of amendments on pollution control through several phytoremediation pathways, their contribution to other soil ecosystem functions supporting different ecosystem services remains sparsely documented. This current review aims at (i) updating the state of the art about the contribution of organic, mineral and microbial amendments in improving phytostabilization, phytoextraction of inorganic and phytodegradation of organic pollutants and (ii) reviewing their potential beneficial effects on soil microbiota, nutrient cycling, plant growth and carbon sequestration. We found that the benefits of amendment application during phytoremediation go beyond limiting the dispersion of pollutants as they enable a more rapid recovery of soil functions leading to wider environmental, social and economic gains. Effects of amendments on plant growth are amendment-specific, and their effect on carbon balance needs more investigation. We also pointed out some research questions that should be investigated to improve amendment-assisted phytoremediation strategies and discussed some perspectives to help phytomanagement projects to improve their economic sustainability
Auguste Mariette. Deux siècles après
No full textInternational audienceA l'occasion du bicentenaire de la naissance d'Auguste Mariette (Boulogne-sur-Mer, 1821 - Le Caire, 1881), nous avons souhaité revenir sur sa personnalité, mais aussi nous interroger sur la place qu'il a tenue dans cette science nouvelle, l'égyptologie, qui en était alors à ses balbutiements, une trentaine d'années après le déchiffrement des hiéroglyphes par un autre Français, Jean-François Champollion.Il existe des biographies d'Auguste Mariette, mais aucun ouvrage ne traite de la place de son oeuvre dans l'histoire politique et culturelle de l'Egypte au XIXe siècle et dans celle des découvertes archéologiques.Ce livre bénéficie d'une approche diversifiée et pluridisciplinaire (archéologues, historiens, conservateurs et muséographie aussi bien français qu'étrangers) et de sources archivistique inédites : vingt-quatre intervenants de différents horizons géographiques et professionnels ont contribué à cet ouvrage en présentant de nouveaux documents ou points de vue
L’accueil du régime de protection et de préservation du milieu marin des détroits internationaux dans le droit français de la police maritime d’urgence.
No full textConsistency analysis of water diffuse attenuation between ICESat-2 and MODIS in Marginal Sea: A case study in China Sea
No full textInternational audienceRecent studies highlight the application of deriving the attenuation coefficient from spaceborne photon-counting lidar ATLAS/ICESat-2 over open oceans on global scales. However, its performance in the more optically complex and variable environments of marginal seas, which are more susceptible to human activity, has not been validated yet. In this study, we present an in-depth analysis of the consistency between diffuse attenuation coefficient (K d ) detection from MODIS and ICESat-2 in China's Marginal Seas. Findings demonstrate that ICESat-2 possesses strong capabilities for the retrieval of the attenuation coefficient across differing aquatic environments. However, discrepancies exist between the lidar system attenuation coefficient obtained from ICESat-2 and the diffuse attenuation coefficient determined by MODIS, influenced by factors such as multiple scattering. Implementation of a novel multiple scattering correction model demonstrates a notable ability in significantly reducing the inconsistency. Validation with in-situ Biogeochemical Argo float measurements reveals an enhancement in the accuracy of lidar-derived diffuse attenuation coefficients upon correction, with the mean absolute percent difference between lidar-derived K d and Argo-K d decreasing from 26 % to 15.7 %. The multiple scattering model developed can bridge the gap between the passive and active remote sensing detection and improve the reliability of lidar-derived attenuation coefficients. Fusing these two missions will greatly improve ocean observation capabilities, providing unprecedented opportunities for precise and comprehensive assessment of marine light environments. This approach has broad implications for ocean science and the application of satellite remote sensing in environmental studies
Surface circulation characterization along the middle southern coastal region of Vietnam from high-frequency radar and numerical modeling
No full textInternational audienceCoastal water dynamics along the Vietnamese middle southern coast (VMSC) region, part of the South China Sea, are highly complex with large spatiotemporal variability whose drivers are not yet well understood. For the first time, high-resolution surface current data from highfrequency radar (HFR) measurements were obtained in this region during the early (transition) phase of the Asian summer monsoon. The data were used for comparison with simulation results from a circulation model, SYMPHONIE, and ultimately to optimize the wind forcing in the model. Both modeling and HFR were able to show the spatial and temporal evolution of the surface circulation, but some discrepancies were found between model and HFR data on some days, coinciding with the evolution of the wind. Two methods were used to optimize the wind forcing, namely the ensemble perturbation smoother (EnPS) and the wind correction method using wind-driven surface currents (EkW). Both methods achieved a significant reduction (∼ 36 %-40 %) in the error of the surface current velocity fields compared to the measured data. Optimized winds obtained from the two methods were compared with satellite wind data for validation. The results show that both optimization methods performed better in the far field, where topography no longer affects the coastal surface circulation. The optimization results revealed that the surface circulation is driven not only by winds but also by other factors such as intrinsic ocean variability, which is not entirely controlled by boundary conditions. This indicates the potential usefulness of large velocity datasets and other data fusion methods to effectively improve modeling results
Diesel generator exhaust emissions : Chemical characterization and cytotoxicity in bladder spheroids
No full textInternational audienceDiesel generators, widely used in developing countries, compensating for long power outages and blackouts, are significant sources of air pollution. In this study, diesel exhaust particulate matter (DEPM) samples were collected from two midsize generators operating in Beirut by cascade impaction, and gravimetrically analyzed for size. The smallest fraction captured on quartz filter was then chemically characterized for its constituents, and examined for its cytotoxicity on 2D and 3D human uroepithelial cell cultures. Results showed that 87 % of collected PM are quasi-ultrafine (<0.33 μm in diameter), and marked high emissions of organic and elemental carbon (OC/EC), elements and metals, particularly Ca, Fe, S, Al, and Ti, and polyaromatic hydrocarbons (PAHs), mainly Benzo[g,h,i]perylene and Dibenzo[a,h]anthracene, as well as high emissions of dioxins, furans and polychlorinated biphenyls (PCBs), particularly OCDD, 1,2,3,4,6,7,8 HpCDF, and PCB118. In addition, in vitro testing showed decreased proliferation, viability, and spheroid formation ability only at high concentrations. In conclusion, DEPM from domestic generators consists of a wide panel of potent toxicants, notably genotoxic, carcinogenic, and endocrine disrupting compounds. Additionally, in vitro results provide a solid basis to further examine the potential contribution of DEPM to bladder tumorigenesis in established cell culture models
IoT-AID: Leveraging XAI for Conversational Recommendations in Cyber-Physical Systems
No full textInternational audienceThe rapid evolution of Industry 4.0 has introduced transformative technologies such as the Internet of Things (IoT), Artificial Intelligence (AI), and big data, facilitating real-time data collection, processing, and decision-making. At the heart of this revolution lies Cyber-Physical Systems (CPS), which integrate computational algorithms with physical components to create intelligent, resilient, and adaptive systems. However, CPS deployment remains complex due to the need for extensive domain expertise. This paper introduces IoT-AID, a novel Explainable AI (XAI)-driven Cyber-Physical Recommendation System (CPRS) that enhances transparency, trust, and efficiency in CPS design. IoT-AID integrates traditional machine learning models, deep learning architectures, and fine-tuned transformer-based models with XAI techniques to automate and improve CPS configuration. Our approach ensures that AI-driven recommendations are interpretable, thereby increasing adoption across industries