Publication Server of Hochschule Darmstadt
Not a member yet
    259 research outputs found

    Human Hand Action Recognition in Industrial Assembly

    No full text
    Manual assembly remains a crucial aspect of industrial production, especially in non-standardized pre-assembly lines where human adaptability to new processes plays a significant role. Despite the human ability to quickly adjust to new tasks, errors frequently occur, particularly at the beginning and end of a task or during transitions between workstations, often due to unfamiliarity or fatigue. This thesis presents the development of an assistance system designed to enhance the accuracy of manual assembly tasks by recognizing and classifying fine-grained human actions through video data. The system guides workers and alerts them to potential errors, including those that might arise when components are placed correctly but improperly, such as a screw inserted without being tightened. Given the lack of industrial-standard training data, a new dataset, Industrial Hand Assembly Dataset V1 (IHADV1), is introduced, containing eleven assembly classes. The use of skeleton-based methods for feature extraction supports efficient and cost-effective training of a spatiotemporal Transformer network. Initial models demonstrated 87% accuracy with a significant reduction in trainable parameters, highlighting the dataset’s efficacy for capturing detailed assembly movements and the value of spatiotemporal analysis in skeletal data. Further advancements in the methodology, including the use of cross-attention mechanisms at the encoder level, resulted in an accuracy of 99%. The thesis also explores self-supervised learning techniques, such as random masking on non-industrial data, leading to a model capable of achieving over 90% accuracy on the fine tuned classes with a significant reduction in labeled data. Additionally a curriculum-based self-learning approach was developed to enable the model to adapt to evolving industrial environments and integrate new assembly classes, ensuring continuous improvement during operational deployment. The findings suggest promising applications for the assistance system in industrial settings, with potential for scalable and self-sustaining advancements in assembly line efficiency

    Structured exploration of attachment history in psychotherapy and counceling

    No full text
    Hintergrund Das Adult Attachment Interview (AAI) erfasst Bindungsrepräsentanzen, die sowohl für die Forschung als auch für die klinische Praxis von hoher Relevanz sind. Als Goldstandard der Erfassung von Bindung mit dem Ziel einer validen Bindungsklassifikation setzt das AAI eine intensive Schulung mit Reliabilitätsprüfung voraus. Ziele der Arbeit Um Bindungserfahrungen und innere Arbeitsmodelle von Bindung in einem zeitlich begrenzten, psychotherapeutischem Kontext systematisch zu explorieren, schlagen wir ein AAI-Protokoll zur klinischen Anwendung vor, ohne den Anspruch einer vollständigen oder reliablen Bindungsklassifikation zu erheben. Material und Methoden Hierzu greifen wir auf ausgewählte Fragen aus dem AAI zurück. Wir stellen Leitfragen vor, mithilfe derer die Antworten von PatientInnen bindungstheoretisch eingeordnet werden können. Das Vorgehen illustrieren wir anhand eines kurzen Fallbeispiels. Darüber hinaus geben wir praxisnahe Hinweise für eine strukturierte Durchführung und benennen typische Fallstricke – beispielsweise im Hinblick auf die Konfrontation mit Widersprüchen oder Interpretation der PatientInnenaussagen während des Interviews. Ergebnisse und Diskussion Die strukturierte Erfassung von Bindungserfahrungen kann TherapeutInnen – unabhängig vom ihrer therapeutischen Ausrichtung – dabei unterstützen, Hypothesen zur Entstehung und zur Aufrechterhaltung psychischer Störungen zu entwickeln. Zugleich liefern die gewonnenen Informationen wertvolle Hinweise zur Beziehungsgestaltung von PatientInnen – sowohl für deren Außenbeziehungen als auch im Hinblick auf die therapeutische Beziehung.Background The adult attachment interview (AAI) assesses attachment representations, which are highly relevant for both research and clinical practice. As the gold standard for measuring attachment, the AAI aims for a valid attachment classification, requiring intensive training with reliability checks. Objectives In order to systematically explore attachment experiences and internal working models of attachment within a time-limited psychotherapeutic setting, we propose an AAI protocol for clinical application without claiming it to be a complete and reliable attachment classification. Material and methods For this purpose, we draw on selected questions from the AAI. We present guiding questions that enable clinicians to comprehend patients’ responses from an attachment-theoretical perspective. The approach is illustrated using a brief case example. In addition, we provide practical guidance for structured implementation and identify typical pitfalls, for example, regarding the confrontation with inconsistencies or the interpretation of patients’ statements during the interview. Results and discussion The structured assessment of attachment experiences can help therapists, regardless of their therapeutic orientation, to develop hypotheses about the development and maintenance of psychological disorders. At the same time, the information obtained provides valuable insights for shaping patients’ relationships, both in their external relationships and with respect to the therapeutic relationship

    Numerical Evaluation and Optimization of the Mechanical Properties of Particle Reinforced Composites

    Get PDF
    In today’s world, composite materials form the bases of many products, ranging from simple’ packaging material to the parts of a wind turbine. Even though computational methods have entered the modern product development cycle, material development has largely stayed unchanged with the use of personal and energy intensive trial and error approaches. This traditional method necessitates the manufacturing of multiple specimens for physical testing, resulting in a long development time while also generating a lot of waste in the process. Numerical methods such as Finite Element Analysis (FEA) in combination with optimization methods can act as an alternative pathway to shrink the development time of a novel composite and reduce the wastage of materials, time and money. The goal of this work was to develop a numerical method for particle reinforced composites that generates microstructures with targeted effective material properties intended for a specific application. These requirements stated during the development process alongside with a set of constraints are utilized for the optimization algorithm to generate an optimized microstructure, drawing upon a data bank for the individual material phases. Digital twins of particles encountered during the research are obtained by use of analytical functions such as Spherical Harmonics, Super Ellipsoids or other equations, which are in the following called ‘exact’ particles. Numerical studies based on FEA were conducted on representative volume elements (RVE) of particle reinforced composites to obtain effective elastic properties The results of the FEA calculations for spherical particles were then compared with results obtained with the use of micromechanical models, such as the Mori-Tanaka scheme, Dilute inclusion. Evaluations of the effect of the particle distribution on the elastic properties of the composite were studied, comparing a homogeneous particle distribution with two distinctly different particle cluster distributions. A numerical surrogate model was developed to approximate the effective elastic properties of the composite with ‘exact’ particles. This method is intended to reduce the computational effort such as calculation time and RAM requirements of evaluating the composites effective elastic material properties in comparison to RVE containing the surrogates’ ‘exact’ particle counterpart. This simplification makes it viable to explore different combinations of particle shapes for different matrix materials. Heuristic optimization methods such as Simulated Annealing, Genetic Algorithm and Particle Swarm Optimization (PSO) were explored for finding an optimal material combination to achieve the targeted effective material properties of the composite. For this a function was derived to obtain the optimal material mixture of the composite, which achieves the targeted effective material properties of the compound. The different heuristic methods were compared according to their numerical stability during optimization and the PSO method was chosen. Numerical methods to generate and evaluate conductive particle reinforced polymer matrix composites were explored, which can utilize a material mixture of two particle shapes such as disc-shaped and line-shaped in a polymer matrix. Lastly the application of machine learning methods such as feed forward neural networks were explored to enable a swift quantification of all the possible solutions with regards to different particle and matrix materials and provide a material mixture which achieves the targeted effective elastic properties

    The Super‐Recogniser Advantage Extends to the Detection of Digitally Manipulated Faces

    Get PDF
    Face recognition by human officials remains the predominant method of identity verification in security‐critical contexts. The integrity of this process can be compromised by sophisticated fraud attacks using manipulated face images. Therefore, in this study, we examine whether human observers can detect digitally manipulated passport photos, and whether super‐recognisers (SRs) outperform typical recogniser controls. Using two face manipulation detection tasks (DFMD1, DFMD2), participants were asked to decide whether a ‘suspected’ passport photo had been digitally manipulated. SRs were found to significantly outperform controls; this effect was not the result of a ‘speed‐accuracy trade‐off’. Individual differences on tests of face identification aptitude, self‐rated ability, and response times, accounted for over 20% of the variance in manipulated image detection sensitivity. Taken together, these findings show that, despite increasing sophistication in digital face manipulation techniques, there is still utility in employing human operators, particularly SRs, to detect them

    Hybride KI mit Machine Learning und Knowledge Graphs

    Get PDF
    Seit den 2010er Jahren hat die Künstliche Intelligenz (KI) durch Erfolge im Machine Learning (ML) einen enormen Schub erfahren. Einerseits durch die stark angewachsene Menge verfügbarer digitaler Daten und andererseits durch Innovationen im Bereich der Künstlichen Neuronalen Netze und des Deep Learning (DL). Wissensbasierte KI umfasst neben traditionellen Expertensystemen und Regelsystemen auch die Technologien und Standards, welche im Rahmen der Semantic Web Initiative seit den 1990er Jahren entwickelt wurden. Sie ermöglichten unter anderem die Entwicklung umfangreicher Knowledge Graphs (Wissensnetze). Hybride KI Ansätze kombinieren Machine Learning und wissensbasierte KI. Da sie als erfolgversprechend gelten, werden sie seit Jahren erforscht. Dieser Sammelband zeigt, wie innovative hybride KI-Verfahren bereits heute erfolgreich in der Praxis eingesetzt werden

    Secure Contactless Fingerprint Recognition

    Get PDF
    Fingerprints, i.e. ridge and valley patterns on the tip of a human finger, are one of the most important biometric characteristics due to their known uniqueness and persistence properties. Large-scale fingerprint recognition systems are not only used worldwide by law enforcement and forensic agencies, they are also deployed in the mobile market and in nationwide applications. In recent years, contactless fingerprint recognition has become a viable alternative to established contact-based methods. The contactless capturing process avoids distinct problems, e.g. signal of low contrast caused by dirt or humidity and left-over latent fingerprints on the capture surface. Moreover, contactless schemes provide a faster and more hygienic as well as a more convenient capturing process and hence have a higher user acceptance. However, contactless fingerprint recognition introduces new challenges. Environmental influences such as an uncontrolled background and varying illumination and an unconstrained finger positioning are especially a challenge for mobile recognition schemes. This Thesis contributes to an efficient and secure mobile contactless fingerprint recognition process. The work addresses various vital aspects along the contactless fingerprint recognition pipeline. The mobile, automatic capturing, segmentation and pre-processing of contactless fingerprint samples represents a central focus of this Thesis. Furthermore, contributions to the topics of quality assessment, feature extraction and presentation attack detection are conducted. To enable new research directions, such as training deep learning-based algorithms, a generator for synthetic mobile contactless fingerprint samples is also suggested. The results proposed in this Thesis show improvements on several components of the recognition method which contribute to an increased biometric performance, security and comfort level. Moreover, challenges and limitations are discussed.Fingerabdrücke, also die Papilarleisten auf der Spitze eines menschlichen Fingers, sind aufgrund ihrer nachgewiesenen Einzigartigkeit und Beständigkeit eines der wichtigsten biometrischen Merkmale. Groß angelegte Fingerabdruckerkennungssysteme werden nicht nur weltweit von Strafverfolgungsbehörden und Forensikern eingesetzt, sondern auch im mobilen Bereich und in überregionalen Anwendungen. In den letzten Jahren hat sich die berührungslose Fingerabdruckerkennung zu einer praktikablen Alternative zu den etablierten kontaktbasierten Verfahren entwickelt. Das berührungslose Erkennungsverfahren vermeidet verschiedene Herausforderungen, wie z.B. Abbildungen mit geringem Kontrast, die durch Schmutz oder Feuchtigkeit verursacht werden, und latente Fingerabdrücke, die auf der Oberfläche des Erkennungsgerätes zurückbleiben. Darüber hinaus bieten kontaktlose Verfahren einen schnelleren und hygienischeren sowie bequemeren Erkennungsprozess und haben daher eine höhere Benutzerakzeptanz. Die kontaktlose Erkennung von Fingerabdrücken bringt jedoch neue Herausforderungen mit sich. Umwelteinflüsse wie ein unkontrollierter Hintergrund und variierende Beleuchtung sowie eine beliebige Positionierung der Finger stellen insbesondere für mobile Erkennungsverfahren eine Herausforderung dar. Diese Arbeit leistet einen Beitrag zu einem effizienten und sicheren mobilen kontaktlosen Fingerabdruckerkennungsverfahren. Die Arbeit befasst sich mit verschiedenen wichtigen Aspekten in der Pipeline der kontaktlosen Fingerabdruckerkennung. Die mobile, automatische Erfassung, Segmentierung und Vorverarbeitung von kontaktlosen Fingerabdruckmustern stellt einen zentralen Schwerpunkt dieser Arbeit dar. Darüber hinaus werden Beiträge zu den Themen Qualitätsbewertung, Merkmalsextraktion und Erkennung von Präsentationsangriffen geleistet. Um neue Forschungsrichtungen zu ermöglichen, wie z.B. das Training von Deep-Learning-basierten Algorithmen, wird auch ein Generator für synthetische mobile kontaktlose Fingerabdruckmuster vorgeschlagen. Die in dieser Arbeit vorgestellten Ergebnisse zeigen Verbesserungen verschiedener Komponenten der Erkennungsmethode, die zu einer verbesserten biometrischen Leistung, Sicherheit und Komfort beitragen. Darüber hinaus werden Herausforderungen und Grenzen diskutiert

    Discovery and Characterization of Novel Non-Hydroxamate HDAC11 Inhibitors

    No full text
    Histone deacetylase 11 (HDAC11), the sole member of class IV HDACs, has gained prominence due to its unique enzymatic profile and pathological relevance in cancer, neurodegenerative, inflammatory diseases, and metabolic disorders. However, only a limited number of selective HDAC11 inhibitors have been identified, and many of these contain a potentially mutagenic hydroxamic acid as a zinc-chelating motif. Consequently, there is an imperative to identify potent and selective non-hydroxamate HDAC11 inhibitors with improved physicochemical properties. In this study, we conducted an extensive experimental high-throughput screening of 10,281 structurally diverse compounds to identify novel HDAC11 inhibitors. Two promising candidates, caffeic acid phenethyl ester (CAPE) and compound 9SPC045H03, both lacking a hydroxamic acid warhead, were discovered, showing micromolar inhibitory potency (IC50 = 1.5 and 2.3 µM, respectively), fast and reversible binding, and remarkable isozyme selectivity. Molecular docking revealed distinct zinc-chelating mechanisms involving either carbonyl oxygen (CAPE) or pyridine nitrogen (9SPC045H03), in contrast to canonical hydroxamates. Both compounds are drug-like and exhibit favorable physicochemical and pharmacokinetic profiles, particularly beneficial water solubility and good adsorption, making them valuable starting points for further optimization. These findings open new avenues for the development of selective, non-hydroxamate HDAC11 inhibitors with potential therapeutic applications

    The Situation Of Professionals In The Public Social Welfare Office Under The Magnifying Glass

    No full text
    Der folgende Beitrag gibt Einblicke in eine im Jahr 2024 in Baden-Württemberg, Hessen und Nordrhein-Westfalen durchgeführte Studie zur Fachkraftsituation im Allgemeinen Sozialen Dienst des Jugendamtes (ASD). Im Fokus stehen neben organisationalen und strukturellen Aspekten vor allem das Belastungserleben der im zentralen Basisdienst Tätigen, aktuelle problematische Entwicklungen sowie Bemühungen im Hinblick auf die Bewältigung der herausfordernden Gesamtsituation und die Gewinnung und Bindung von Personal. Dieser Beitrag ist Teil des Schwerpunktes „Fachkräftemangel und -bedarf in der Kinder- und Jugendhilfe – Analysen, Einordnungen und Impulse insbesondere zur Situation im zentralen Basisdienst (ASD) des Jugendamtes“ in Ausgabe 5/25.The following article provides insights into findings from research in 2024 in Baden-Württemberg, Hesse and North Rhine-Westphalia on the situation of professionals in public youth welfare offices (ASD). Beneath organizational and structural aspects, experiences and perceptions of social work professionals in the context of current problematic developments as well as reports on efforts with regard to coping with the challenging situation and recruiting and retaining staff are addressed

    Catalysis in Silver Nanocube Formation: The Role of Iron Ions in Non-Polar Solvents

    No full text
    Plasmonics is a rapidly growing field of research based on plasmonic nanostructures. To exploit the full potential of this fascinating class of materials, it is indispensable to tune and optimize the properties of these structures, which requires precise knowledge and optimization of their synthesis processes. Plasmonic silver nanocubes for applications in nonpolar media are obtained by an AgCl-mediated hot-injection method. In this process, catalysis by Fe species is of central importance, as the Fe species influence the reaction in multiple ways, enabling a finely balanced control of the nanocube synthesis. Using electron microscopy, optical spectroscopy, and X-ray photoelectron spectroscopy, it is shown that the Fe species not only direct the reaction of the Ag precursor to the formation of AgCl nanoparticles instead of icosahedral Ag nanoparticles but also enhance the reduction rate of AgCl, from which the Ag nanocubes are formed and grow. Based on these results, a detailed reaction mechanism is proposed. An additional comparison of the effects of different metal ions on the reaction shows that iron ions are highly likely to be specific as catalysts for this synthesis. The results also indicate that the Fe ions are likely present in the form of an organic iron complex, catalyzing the chloride transfer

    Künstliche Intelligenz im Studium - Eine quantitative Längsschnittstudie zur Nutzung KI-basierter Tools durch Studierende

    No full text
    KI-basierte Tools wie ChatGPT und GPT-4 sind fester Bestandteil des Hochschulalltags und verändern die Art und Weise, wie Studierende lernen und arbeiten. Vor diesem Hintergrund wurde eine deutschlandweite quantitative Studie durchgeführt, um die Nutzung und die damit verbundenen Präferenzen der Studieren-den zu analysieren. Ziel der Untersuchung ist es, ein fundiertes Verständnis über die Verbreitung, Intensität und Einsatzbereiche KI-gestützter Tools im Studium zu gewin-nen. An der aktuellen Befragung haben sich 4910 Studierende beteiligt. Die Ergebnisse zei-gen, dass mittlerweile mehr als 90 % der befragten Studierenden KI-basierte Tools im Studium nutzen – ein deutlicher Anstieg im Vergleich zur Erhebung von 2023, in der der Anteil bei 63 % lag. Besonders häufig kommen ChatGPT und DeepL zum Einsatz, wobei ChatGPT sowohl in der kostenfreien als auch in der kostenpflichtigen Version mit Abstand das meistgenutzte Tool ist. Die differenzierte Analyse des Nutzungsverhaltens verdeutlicht, dass KI-basierte Tools zunehmend in verschiedenen akademischen Kontexten eingesetzt werden. Die Klärung von Verständnisfragen und die Erklärung fachspezifischer Konzepte stellen weiterhin die häufigsten Anwendungsfälle dar. Darüber hinaus zeigt sich ein signifikanter Anstieg in der Nutzung für Recherchen, die Erstellung wissenschaftlicher Texte, Übersetzungen sowie datengetriebene Analysen. Insbesondere Studierende der Ingenieurwissenschaf-ten greifen verstärkt auf KI-gestützte Werkzeuge zurück. Die Ergebnisse unterstreichen, dass KI-basierte Tools innerhalb kurzer Zeit zu einem integralen Bestandteil des Studiums geworden sind und sich ihr Einsatzbereich konti-nuierlich erweitert

    211

    full texts

    259

    metadata records
    Updated in last 30 days.
    Publication Server of Hochschule Darmstadt
    Access Repository Dashboard
    Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇