Portail des publications scientifiques IMT Mines Alès
Not a member yet
5198 research outputs found
Sort by
Postural risk assessment and motion capture to improve the ergonomics of a workstation
No full textNational audienceThe World Health Organization (WHO) indicates that Musculoskeletal Disorders (MSD) “range from those that arise suddenly and are short-lived, such as fractures, sprains, and strains, tolifelong conditions associated with ongoing pain and disability” [1]. These conditions can man-ifest in individuals of any age and across global populations. They carry significant economicimplications, leading to a decline in work efficiency and adversely impacting the well-being ofthose affected.The "Regional Occupational Health Plan 2022-2025" for the Occitanie region mentions sev-eral figures to illustrate the necessity of placing health and prevention at the heart of workin the region: MSD represent 80% of recognized occupational diseases; 71% of employees areexposed to postural and articular constraints; 32% of employees are exposed to one or morechemicals regardless of their sector of activity. As for the statistical service responsible forlabor, employment, and vocational training within the public statistical system (DARES), itsanalysis No. 031 from August 2022 lists the thresholds above which exposure can be qualifiedas "onerous" (according to the Medical Surveillance of Employees’ Exposure to OccupationalRisks (SUMER) survey from 2017). However, it specifies that most quantification methods arebased on medical investigators’ expertise and employees’ perceptions [2].We focus on identifying and automatically calculating ergonomic risks in this context andthis preliminary work. First, we conduct a literature review on a well-known method called“REBA". Then, we use motion capture to analyze the human motion and calculate the REBAscore automaticall
Concentrated O/W Emulsion Stability of Non-Ionic Chitosan Oligomer Surfactants Modified by Epoxidized Fatty Chains at pH7: Influence of Emulsification Conditions
No full textInternational audienceIn this study, chitosan-based surfactants were synthesized by epoxy–amine chemistry to stabilize concentrated O/W emulsions at pH7. Chitosan was first depolymerized by nitrous deamination to obtain chitooligosaccharides (COS) with degrees of polymerization of 10 (DP10) and 20 (DP20). Then, three different epoxidized fatty chains, i.e., octyl/decyl glycidyl ether (C9), hexadecyl glycidyl ether (C16) and epoxidized cardanol (card), were grafted onto the amine groups of chitosan to form six amphiphilic structures. NMR measurements revealed grafting efficiencies ranging from 1 to 30% while HLB values ranged from 13 to 20. The relationships between these surfactant structures and their adsorption properties were investigated by tensiometric measurements, highlighting the need for a short hydrophilic moiety and high grafting efficiency to obtain the best adsorption. Subsequently, concentrated O/W emulsions (66% of oil) at pH7 were produced using COS-based surfactants and the impact of stirring time and speed during the emulsification process was described through rheological, droplet size and microscopy measurements. Finally, emulsions were stored over 2 months in order to study the destabilization phenomenon into the mixture, i.e., coalescence and creaming, by using laser granulometry and Turbiscan. Results demonstrated that stability could be enhanced by increasing emulsion viscosity, reducing droplet size or optimizing the adsorption layer at the O/W interfaces
Uranium and thorium recovery from naturally contaminated solutions using stable functionalized sorbents
No full textInternational audienc
INGEsT: An Open-Source Behavioral Setup for Studying Self-motivated Ingestive Behavior and Learned Operant Behavior
No full textIngestive behavior is driven by negative internal hunger and thirst states, as well as by positive expected rewards. Although the neural substrates underlying feeding and drinking behaviors have been widely investigated, they have primarily been studied in isolation, even though eating can also trigger thirst, and vice versa. Thus, it is still unclear how the brain encodes body states, recalls the memory of food and water reward outcomes, generates feeding/drinking motivation, and triggers ingestive behavior. Here, we developed an INstrument for Gauging Eating and Thirst (INGEsT), a custom-made behavioral chamber which allows for precise measurement of both feeding and drinking by combining a FED3 food dispenser, lickometers for dispensing liquid, a camera for behavioral tracking, LED light for optogenetics, and calcium imaging miniscope. In addition, in vivo calcium imaging, optogenetics, and video recordings are well synchronized with animal behaviors, e.g., nose pokes, pellet retrieval, and water licking, by using a Bpod microprocessor and timestamping behavioral and imaging data. The INGEsT behavioral chamber enables many types of experiments, including free feeding/drinking, operant behavior to obtain food or water, and food/water choice behavior. Here, we tracked activity of insular cortex and mPFC Htr3a neurons using miniscopes and demonstrate that these neurons encode many aspects of ingestive behavior during operant learning and food/water choice and that their activity can be tuned by internal state. Overall, we have built a platform, consisting of both hardware and software, to precisely monitor innate ingestive, and learned operant, behaviors and to investigate the neural correlates of self-motivated and learned feeding/drinking behaviors
Flame Retardancy of Short Flax Fibers Modified by Radiation-Induced Grafting of Phosphonated Monomers: Comparison between Pre- and Simultaneous Irradiation Grafting
No full textInternational audienceShort flax fibers have been modified by radiation-induced grafting using methacrylate monomers containing phosphorus to give them a flame-retardant character. Two methodologies, namely pre-irradiation and simultaneous irradiation grafting, were examined. Certain parameters, notably the dose and the irradiation source (e-Beam and γ rays), were evaluated. The grafting efficiency, in terms of phosphorus content (mass percentage), was measured by X-ray fluorescence spectrometry (XRF). Using simultaneous irradiation, 2.39 wt% phosphorus could be obtained from 10 kGy, compared to 100 kGy in pre-irradiation. Furthermore, for similar phosphorus levels, the location of the grafted polymer chains was different for the two methodologies. The effect of phosphorus content on thermal properties and fire behavior was evaluated on a microscopic scale using a pyrolytic flow combustion calorimeter (PCFC) and on a laboratory scale using a cone calorimeter. It was then pointed out that flammability was linked to the phosphorus content and likely its location, which is associated with the radiation-induced grafting methodology, showing that the grafting conditions influence the final fire properties. Simultaneous irradiation, thus, proved to be more interesting in terms of efficiency and final properties
Pignistic-Plausibility Interval-Criterion and Pignistic-Strong Dominance Convex Mixture Criterion Based Evidential Set-Valued Classifications
No full textAbstractImperfect data have to be processed in a different way when they are involved in classification or regression tasks. For instance, in the case of supervised classification, predicting a subset of candidate classes when imperfect data are present is preferable to predicting a single class using point prediction methods which offer less guarantee. This is even more necessary when machine learning algorithms are involved in sensitive domains, e.g., medical field. Of course, the corresponding classifier should be cautious, i.e., the predicted subset of candidate classes contains the true class, but also relevant or precise, i.e., its size is not too large. The well known Strong Dominance based classification algorithm (SD) is a good candidate as a robust method but some times the predicted subsets are too large. This paper proposes two cautious classification algorithms within the framework of belief functions that aim to make flexible the SD classifier. Indeed, these classifiers control the granularity of the partial order that is returned by Strong Dominance relation. Ideally, an output which is a compromise between the robustness, i.e., cautiousness, offered by the SD and the precision offered by point prediction classifiers must be guaranteed for the decision-maker regarding his preferences. The first classifier is based on a interval criterion that is built from the pignistic criterion to which is associated a threshold while the second classifier is based on a convex mixture between the pignistic criterion and the Strong Dominance binary relation. The outputs of these three classifiers are theoretically and experimentaly compared
Le test d’indentation instrumentée, une méthode adaptée pour mesurer la tendreté de la viande ?
No full textInternational audienceThe « beef » sector is exploring how to guarantee delivering products to its clients with regular quality. However, the methods used toqualify beef tenderness are mainly invasive since they require sampling meat samples to be qualified. Therefore, tenderness is little or not atall evaluated by the sector. The objective of this study was to propose a non-invasive test, which ideally can be performed on a carcass and/oron-line, replacing the invasive tests used previously in laboratories (Warner-Bratzler shearing test, hardness test, penetration test, baking test).The new test is an indentation/relaxation/recovery test that leads to an estimation of tenderness indicators. The indentation method consists inthe penetration of the meat with a hemispherical point, maintaining it in position then its rapid withdrawal, a phase associated with the returnto equilibrium during which the movement at the meat surface is closely observed. A preliminary study on two pieces of meat with very different tenderness levels associates some of these indicators with data from compression tests (tensiometer), a device made available for thisstudy by the Association for Development of the Meat Institute in Clermont-Ferrand (ADIV). In particular, after indentation, the gite roundreturns more quickly to its equilibrium state than the sirloin steak. The tendency that was observed on this “recovery” stage seems to becorrelated with data obtained during penetrometry trials on similar samples. Statistical studies now need to be performed to validate the use ofthis tool for the industrial qualification of pieces of meat.La filière « viande » réfléchit à assurer la garantie de délivrer à ses clients des produits réguliers en qualité. Toutefois, les méthodes utiliséespour qualifier la tendreté de la viande sont pour la majorité destructives puisqu’elles nécessitent le prélèvement d’échantillons de la viande àqualifier. De ce fait, la tendreté n’est pas ou peu évaluée dans la filière. L'objectif de cette étude est de proposer un test non destructif, idéalementréalisable sur carcasse et/ou en ligne permettant de se substituer aux tests invasifs classiquement utilisés au laboratoire (test de cisaillement deWarner-Bratzler, test de dureté, test de pénétration, test de cuisson). L’essai envisagé est un test d’indentation/relaxation/recouvrance quiconduit à l’estimation d’indicateurs de tendreté. Le protocole d’indentation consiste en la pénétration d’une pointe hémisphérique, en sonmaintien en position et en son rapide retrait, phase liée au retour à l’équilibre, pendant laquelle le mouvement de la surface de la viande estparticulièrement suivi. Un travail préliminaire sur deux pièces de viande dont les niveaux de tendreté sont très différents a permis de mettre enrelation certains de ces indicateurs avec les données issues d’un test par compression (tendéromètre), dispositif mis à disposition pour cetteétude par l’Association pour le Développement de l’Institut de la Viande de Clermont-Ferrand (ADIV). En particulier, après indentation, lerond de gite retourne plus rapidement à son état d’équilibre que le faux filet. La tendance observée sur cette phase dite de « recouvrance »semble corrélée avec des données plus classiques obtenues lors d’essais de pénétromètrie sur des échantillons similaires. Des études statistiquesdoivent maintenant être réalisées pour valider l’utilisation de cet outil pour la qualification industrielle des pièces de viande
De-sync: disruption of synchronization as a key factor in individual and collective creative processes
No full textInternational audienceAbstract Creativity is a key skill for the twenty-first century, where the individual and collective imperative to adapt is omnipresent. Yet, it is still unclear how to put creativity theories into practice, which signals a lacuna in our understanding of the pragmatic means by which we get creative. This paper starts from the identification of a number of gaps in the literature. In particular, individual and group creativity are usually treated separately, and the emphasis on the search for novelty seems to overshadow the importance experts give to the disruption of their habitual patterns of behavior. To overcome these gaps, we propose foundations for a unifying framework that takes the perspective of dynamical systems. Specifically, we suggest that de-synchronization, a hallmark of disruption, is an integral part of the creative processes that operate across individual and collective levels of analysis. We show that by conjuring uncertainty, de-synchronized states provide opportunities for creative reorganization. In order to ground this framework, we survey and discuss existing literature, and focus on group improvisation practices (in particular, music and dance improvisation), where partners use the dynamics of their interaction to bring forth a collective performance in real-time. In these practices, disruption by de-synchronization, termed here as ‘problematization of coordination’, is a pragmatic approach used to push the creative process forward. We suggest that this approach might also be relevant in other types of individual and collective creative processes
FORT-RAJ: A Hybrid Fisheye Model for Real-Time Pedestrian Trajectory Prediction
No full textInternational audienceThis paper introduces FORT-RAJ, a hybrid model designed forpedestrian trajectory prediction in the context of top-view fisheye images. Toachieve this, FORT-RAJ merges the FORT (Fisheye Online Realtime Tracking)algorithm, which tracks people using fisheye cameras without predictioncapabilities, with the GATraj model, known for trajectory prediction but notyet adapted for fisheye images. The proposed method, FORT-RAJ, is designedto detect pedestrians, track their trajectories, and predict their futurepositions. It leverages the wide field of view of fisheye cameras whileaddressing the distortions inherent in such images. The experimentsdemonstrated that the FORT-RAJ model performs satisfactorily on fisheyeimages, achieving an Average Displacement Error (ADE) of 0.38 meters andan Final Displacement Error (FDE) of 0.42 meters
