HAL Portal UPHF (Université Polytechnique Hauts-de-France)
Not a member yet
28473 research outputs found
Sort by
Impacts économico-touristiques des méga-événements sportifs à différentes échelles géographiques: Le cas des Jeux Olympiques de Paris 2024 d'un point de vue départemental et régional
No full textInternational audienceAu cours de ces dernières années, la France et la région Hautsde-France ont organisé plusieurs événements sportifs majeurs ouméga-événements comme la finale de la Coupe Davis, des matchs del’Euro 2016, de la Coupe du Monde de Rugby 2023 ou encore desépreuves pour les Jeux Olympiques de Paris 2024.L’objectif de la thèse sera de démontrer, par des régressionséconométrique, l’influence des méga-événements sur le tourisme àdifférentes échelles et son incidence économique par le conceptthéorique du « Tourism Led Growth Hypothesis ». Pour cela, uneétude sera faite à différentes échelles géographiques sur la base d’unmodèle gravitaire. Le premier étudiera l’influence des JeuxOlympiques sur l’arrivée de visiteurs provenant des différentesrégions françaises, par le biais de différents facteurs propres aumodèle gravitaire en plus de facteurs propres aux régions (lienshistoriques notamment, PIB, population). Le second montrera lesimpacts des grands événements sportifs dans différents pays hôtes,avec pour but de déterminer et de comparer l’impact de cesévénements dans ces destinations.La présentation mettra l’accent sur l’avancée des différentsprojets, mais également sur les recommandations potentielles qui vontêtre apportés aux différents décideurs (profils de visiteurs visés pourdes régions parfois en manque d’attractivité comme les Hauts-de-France, mise en valeur de l’événement organisé, conseils pourl’organisation future de méga-événements)
PViTGAtt-IP: Severity Quantification of Lung Infections in Chest X-rays and CT Scans via Parallel and Cross-Attended Encoders
No full textInternational audienceThe development of a robust and adaptive deep learning technique for the diagnosis of pneumonia and the assessment of its severity was a major challenge. Indeed, both chest X-rays (CXR) and CT scans have been widely studied for the diagnosis, detection and quantification of pneumonia. In this paper, a novel approach (PViTGAtt-IP) based on a parallel array of vision transformers is presented, in which the input image is divided into regions of interest. Each region is fed into an individual model and the collective output gives the severity score. Three parallel architectures were also derived and tested. The proposed models were subjected to rigorous tests on two different datasets: RALO CXRs and Per COVID-19 CT scans. The experimental results showed that the proposed models exhibited high performance in accurately predicting scores for both datasets. In particular, the parallel transformers with multi-gate attention proved to be the best performing model. Furthermore, a comparative analysis using state-of-the-art methods showed that our proposed approach consistently achieved competitive or even better performance in terms of the Mean Absolute Error (MAE) and the Pearson Correlation Coefficient (PC). This emphasizes the effectiveness and superiority of our models in the context of diagnosing and assessing the severity of pneumonia. The source codes used in our study are publicly available and can be found at https://github.com/bouthainas/PViTGAtt-IP
Modeling Mechanical Micro-instabilities in Biophysics and Materials Science
No full textInternational audienceFrom biophysics to materials science, mechanical micro-instabilities play a central role in a multitude of materials and structures (adhesion, fracture, friction, phase transformations and so on). Although these systems are very different in length-scale, time-scale and morphology, they are all composed of a number of units that exhibit two or more equilibrium states and can switch from one to another. Schematically, we can identify two main classes of micro-instabilities in these multi-stable systems. On the one hand, we may observe a bi-stable behavior between one ground state and one metastable state, being these two states different, yet mechanically resistant conformations. For instance, in this class can be inscribed conformational (folded to unfolded) transitions in polymers or macromolecules (mostly proteins), and martensitic phase transformations in metallic alloys and nanowires. On the other hand, the second class corresponds to transitions between unbroken and broken states of breakable units of the system. This process can be reversible, partly reversible or irreversible according to the specific physical phenomenon. Examples of this scheme include unzipping of DNA or RNA hairpins, denaturation of macromolecules, biological adhesion, peeling of films in nanotechnology, and fracture phenomena in mechanics of materials. In this review, we offer a description of phenomena related to these mechanical micro-instabilities, ranging from biophysics to materials science. After presenting an overview of typical examples in both contexts, physical-mathematical approaches suitable for modeling and predicting the behavior of these systems will be introduced. In order to consider both thermal and mechanical effects, they are based on the combination of statistical mechanics and micromechanics
‘Catch and Release’ Polymeric Fibers: Versatile Interfaces for Engineering Reversible Platforms for Biomolecular Immobilization and Antibacterial Coatings
No full textInternational audienceABSTRACT Stimuli‐responsive polymeric fibers have emerged as an indispensable material for numerous biomedical applications. Strategies to conjugate functional molecules with high specificity onto these nanofibers are vital to tailor these materials for specific applications. When the functionalization is reversible, these materials can serve as a ‘catch and release’ platform, which widens their applicability. Herein, polymeric fibers with an average diameter of about 237 ± 44 nm, amenable to reversible conjugation, are fabricated using electrospinning. The thiol‐disulfide exchange reaction is employed to functionalize the electrospun fibers with thiol‐containing functional molecules ranging from fluorescent dyes to bioactive ligands for protein immobilization. It is demonstrated that the linked (bio)molecules can be efficiently released in the presence of a thiol‐containing reducing agent. Specifically, pyridyl disulfide (PDS)‐containing copolymers are synthesized using a thiol‐reactive PDS‐based monomer, methyl methacrylate, and poly(ethylene glycol) methacrylate, where the monomers enable thiol‐based specific functionalization, stable fiber formation, and anti‐biofouling characteristics, respectively. After demonstrating efficient functionalization and release using fluorescent dyes and bioactive ligands, these fibers are conjugated with a thiol‐containing cationic antibacterial peptide. It is demonstrated that the released peptide preserves its antibacterial activity against planktonic bacteria as well as biofilms. One can envision that the facile fabrication, efficient functionalization, and on‐demand release attribute of these reversibly functionalizable polymeric fibers disclosed here would be attractive platforms for a wide range of biomedical applications
Improving electrical contact of vertically aligned carbon nanotubes with bonding metals in thermocompression bonding through oxygen plasma treatment
No full textInternational audienceVertically aligned carbon nanotubes (VACNTs) are being widely studied due to their exceptional mechanical, electrical, and thermal properties. The application of thermocompression bonding (TCB) technique has positioned VACNTs as promising candidates for 3D heterogeneous integration, interconnects, and nanopackaging. However, challenges arise from a CNT-cluster layer covering the top surface of the as-grown VACNTs, adversely affecting wettability and electrical contact between the VACNTs and the bonding metals used in the TCB process. To mitigate this issue, we propose using oxygen plasma reactive ion etching (RIE) to treat the top and/or bottom surfaces of the VACNTs. Scanning electron microscopy and Raman spectroscopy characterizations confirm the successful removal of the CNT-cluster layer from the top surface following the RIE treatment, and excessive RIE treatment may cause partial damage to the graphitic structures of the VACNT surface. Direct-current measurements on fabricated VACNT wall π-bridge structures further validate the enhancement in electrical contact between the top surface-treated VACNTs and the bonding metals. This improvement is attributed to the effective removal of the CNT-cluster layer and the relative minimal damage to the graphitic structures, particularly in comparison to the bottom surface-treated VACNTs
An optimal data analysis for real-time body motion control in a perturbed environment
No full textInternational audienceAn optimal data analysis for real-time body motion control in a perturbed environmen
Design and Optimisation of a Vibrating Wing Insect-Size Air Vehicle with Lumped Parameter Models and Compliant Links
No full textInternational audienceAbstract This article presents the design of a microfabricated bio-inspired flapping-wing Nnano Aaerial Vvehicle (NAV), driven by an electromagnetic system. Our approach is based on artificial wings composed of rigid bodies connected by compliant links, which optimise aerodynamic forces though replicating the complex wing kinematics of insects. The originality of this article lies in a new design methodology based on a triple equivalence between a 3D model, a multibody model, and a mass/spring model (0D) which reduces the number of parameters in the problem. This approach facilitates NAV optimisation by using only the mass/spring model, thereby simplifying the design process while maintaining high accuracy. Two wing geometries are studied and optimised in this article to produce large-amplitude wing motions (approximately ), and enabling flapping and twisting motion in quadrature. The results are validated thanks to experimental measurements for the large amplitude and through finite element simulations for the combined motion, confirming the effectiveness of this strategy for a NAV weighing less than 40 mg with a wingspan of under 3 cm
Integrating perceptual quality analysis and caption-based features for robust deepfake video detection
No full textdata on github https://github.com/HessenUPHF/perceptual-caption-detectionInternational audienceInspired by the benefits of both linguistic representations as well as visual quality analysis, we propose a multimodal framework for deepfake detection with reliable accuracy. Our method acquires low-level perceptual features from frames of video, including contrast, brightness, and sharpness, and computes artifact scores for artificial anomaly detection. In parallel, our method produces descriptive frame-level captions, summed up for generating video-level summaries for capturing contextual coherence. For training the model, our method leverages the FaceForensics++ dataset, comprising several techniques of deepFake manipulation, including DeepFake, Face2Face, FaceSwap, as well as NeuralTextures. Metadata comprised of quality measurements, the artifact scores, as well as textual captions, is tokenized, processed by a DeepSeek V2 Lite model, fine-tuned by the Low-Rank Adaptation (LoRA) procedure, as the backbone for classification. We test the proposed model on both intra-dataset as well as cross-dataset settings, demonstrating good accuracy as well as impressive generalization ability. Our method achieves a 96.51% accuracy for classifying, by integrating low-level perceptual as well as high-level semantic reason, effectively marking manipulated content as well as upholding the integrity of the medium in a world of increasingly evolving generative methods. The employed source code as well as dataset preprocessing scripts for our work are made available for public access on: https://github.com/HessenUPHF/perceptual-caption-detectio
Antifungal and phytotoxicity of wood vinegar from biomass waste against Fusarium oxysporum f. sp. cubense TR4 infecting banana plants
No full textInternational audienceFusarium oxysporum f. sp. cubense Tropical Race 4 (Foc TR4) infection poses a serious threat to banana cultivation, leading to significant economic losses due to the lack of effective and sustainable disease management strategies. Wood vinegar, derived from biomass waste, offers a potential eco-friendly alternative to chemical pesticides for disease control. This study evaluated the antifungal activity of wood vinegar from coconut shells and Acacia wood residues at different concentrations (1, 2, 4, 6, 8, and 10% v/v) against Foc TR4 in vitro using the poisoned food technique and assessed its phytotoxicity on banana plants. Our results showed that wood vinegar from both sources completely inhibited the growth of Foc TR4 in vitro at all concentrations tested. The phytotoxicity effects of the different types of wood vinegar on the banana plants were also evaluated using the same concentrations via foliar spray and soil-drenching techniques. Growth parameters, including plant height, stem diameter and leaf number, were recorded at 24, 31, and 39 days after transplanting (DAT). Treatments with lower concentrations (2% v/v) promoted the growth of the banana plants, while at higher concentrations, severe phytotoxicity symptoms such as leaf necrosis and yellowing were observed. These findings provide valuable insights into the potential use of biomass waste-based wood vinegar for controlling Fusarium wilt in bananas
Proposition d’une méthodologie de coopération humain-robot : application à une cellule de désassemblage robotisée
No full textInternational audienceDans le domaine du désassemblage d’appareils électrotechniques, la coopération humain-robot est devenue essentielle. Si les systèmes autonomes progressent, la diversité des produits rend le travail manuel nécessaire. Dès lors, il est crucial de mettre en place une coopération efficiente entre humains et robots. Celle-ci repose sur l’adaptabilité humaine, l’efficacité du robot et une communication entre eux pour anticiper les tâches de désassemblage, tout en priorisant la prise de décision par l’humain et les interactions entre les agents