Eldorado - Repositorium der TU Dortmund

Ressourcen aus und für Forschung, Lehre und Studium

Bei diesem Service handelt es sich um das Institutionelle Repositorium der Technischen Universität Dortmund. Hier werden Ressourcen aus und für Lehre, Studium und Forschung gespeichert, erschlossen und der Öffentlichkeit zugänglich gemacht.

Hauptbereiche in Eldorado

Wählen Sie einen Bereich, um dessen Inhalt anzusehen.

Gerade angezeigt 1 - 5 von 9

Fach- und Universitätsübergreifendes 1319
Fakultäten 13866
Graduate School und Graduiertenkollegs 12
Hochschulverwaltung 1249
Sonderforschungsbereiche 1595

Aktuellste Veröffentlichungen

Generation of training data for handwritten text recognition using latent diffusion models
(2026) Brandenbusch, Kai Ingo Wilhelm; Fink, Gernot A.; Harmeling, Stefan
Handwritten documents have served as the predominant medium for the preservation and transmission of information. Numerous works aim at implementing automatic searchability and information extraction from such documents. Additionally, a research field has emerged that deals with the generation of handwritten words and documents. The task of handwritten text generation (HTG) constitutes the generation of a realistic looking image depicting a given string in a handwriting with a desired style. In particular, the correctness and readability of the generated word as well as the imitation of the desired style pose major challenges. Generative models such as generative adversarial networks and diffusion models have been adopted to approach this task. HTG offers the promising opportunity to generate annotated training data for other document analysis models. This is particularly interesting when training data is generated for a new target dataset for which no annotated examples are available. In this thesis, an HTG system based on a latent diffusion model for the generation of training data for handwritten text recognition models is proposed. In order to generate images for an unseen target dataset, a pretrained masked autoencoder is used to extract style encodings from a set of example images. Together with embeddings of the string to be generated, these encodings are used to condition the generation process using classifier-free guidance. In order to enhance the generation quality for styles from the target dataset, two semi-supervised training schemes for the HTG model are presented in this work. These training schemes enable the model to leverage information about new styles either from examples that are only annotated with writer IDs or from examples without any annotation. The obtained HTG system is used to generate a synthetic dataset which contains samples with handwriting styles similar to those in the target dataset. A handwriting recognition model is then trained on this stylized synthetic dataset. The experimental results demonstrate the successful application of the proposed HTG model for the generation of training data for a handwriting recognition model. Even if the HTG model is trained with a dataset other than the target dataset, it is shown that a recognition model can successfully be trained using only generated training samples. Furthermore, the experiments demonstrate that including unlabeled samples from the target dataset using the proposed semi-supervised training schemes results in considerable improvements of the recognition model trained on the generated data. In summary, the HTG system presented in this thesis offers a promising approach toward the generation of training data for unseen datasets and can facilitate the training of other document analysis models.
Chiral Pd2L4 capsules from readily accessible Tröger’s base ligands inducing circular dichroism on fullerenes C60 and C70
(Cambridge Crystallographic Data Centre, 2024-12-03) Benchimol, Elie; O’Connor, Helen M.; Schmidt, Björn; Bogo, Nicola; Holstein, Julian J.; Lovitt, June I.; Shanmugaraju, Sankarasekaran; Stein, Christopher J.; Gunnlaugsson, Thorfinnur; Clever, Guido H.
The induction of chirality on pristine fullerenes through non-covalent embedding in an asymmetric nano-confinement has only been rarely reported. Bringing molecules with such a unique electronic structure and broad application range into a chiral environment is particularly appealing for the development of chiroptical materials, enantioselective photoredox catalysts and systems showing chirality-induced spin selectivity (CISS). In this study, we report the formation of a chiral, configurationally stable Pd2L4 capsule assembled from a C2-symmetric, ‘ribbon-shaped’ ligand with a Tröger's base naphthalimide (TbNaps) backbone, easily synthesized in three steps from commercially available compounds. Embedding chirality directly into the ligand backbone ensures a relatively lightweight receptor design whose aromatic panels create a strongly shielded inner cavity of about 700 Å3 volume. Fullerenes C60 and C70, as well as a pair of corannulenes, can be bound in acetonitrile (where unsubstituted fullerenes are insoluble) and X-ray structures of host-guest complexes were obtained. Tight interactions between the chiral host and the fullerene guests leads to the induction of a circular dichroism (CD) on the characteristic absorption bands of the forbidden π–π* transitions of the fullerenes, backed up by sTDA TD-DFT calculations and detailed investigation of the electronic excited states.
Case study: flipped classroom with gamification in a hybrid fluid mechanics course
(Wiley, 2024-07-05) Boettcher, Konrad E. R.; Fischer, Michael‐David; Hellmich, Justus
A fluid mechanics course in process engineering designed according to the students’ wishes mixes lecture and exercise in co-teaching and flipped-classroom elements with gamification. The effectiveness increases as the number of points achieved in the exam increases by 31.9 % compared to the average of the four previous years. The withdrawal rate drops from 54.8 % to 17.4 % and the failure rate from 38.7 % to 23.7 %, which enhances efficiency. The students’ self-report shows a better preparation to the course sessions, but they do not feel stressed much additionally. About 80 % of the initially attending students participate at the end of the course.
Essays on regional and public economics
(2026) Büscher, Tobias; Zudenkova, Galina; Riedel, Nadine
This dissertation examines how public spending is allocated to private firms and how fiscal incentives shape this allocation, with a particular focus on spatial patterns. While public economics traditionally emphasizes the size and financing of government spending, comparatively little attention has been paid to how funds are distributed across firms and regions. This dissertation addresses this gap by analyzing two major channels of public spending: public procurement and R&D subsidies. The first chapter studies spatial allocation patterns in European public procurement and documents a home bias. Using a trilateral gravity framework that incorporates the contracting authority, the winning firm, and the place of performance, the analysis shows that from the firm-perspective, proximity to the contracting authority is more important than proximity to the place of performance. This suggests that institutional proximity, social ties, or local favoritism play a larger role than standard trade or transport costs, indicating limited market integration in EU procurement. The second chapter investigates a potential mechanism underlying this localization by focusing on local business taxation in Germany. It shows that municipalities have fiscal incentives to award contracts to local firms, as this increases their tax base. Using a control function approach based on exogenous variation from the German census shock, the results demonstrate that higher local business tax rates lead to a significantly higher share of locally awarded contracts. The third chapter analyzes the effectiveness of public R&D funding at the firm level. Combining administrative funding data with firm-level data, it uses matching on the firm-level and the cluster level. It finds positive direct and indirect effects on fixed assets, the number of employees and quality-adjusted patents. In particular, the results highlight substantial spillover effects within regional clusters, which exceed the direct effects on subsidized firms. Overall, the dissertation provides new empirical evidence on the allocation and effectiveness of public spending. It highlights the importance of local fiscal incentives and institutional factors in shaping allocation decisions and points to potential trade-offs between decentralized policy-making and broader objectives such as market integration and efficiency.
Biodiesel as a sustainable platform chemical enabled by selective partial hydrogenation
(Wiley, 2024-02-20) Roth, Thomas F. H.; Kühl, Alexander; Spiekermann, Maximilian L.; Wegener, Hannes W.; Seidensticker, Thomas
The hydrogenation of polyunsaturated fatty acids (PUFAs) in vegetable oils and their derivatives is essential for their use in many areas, such as biofuels and food chemistry. However, no attempts have been made to adapt this technology to the requirements of further chemical utilization of fatty acid methyl esters as molecular building blocks, especially for particularly promising double-bond reactions. In this work, we, therefore, use three homogeneous catalytic model reactions (hydroformylation, isomerizing methoxycarbonylation, and ethenolysis) to show, firstly, that it is already known from the literature that high PUFA contents have a negative impact on activity and selectivity. Subsequently, using the example of soybean and canola biodiesel, we demonstrate that these key figures can be drastically improved by a preceding selective partial hydrogenation. This makes it possible to first reduce the share of PUFAs to <1 w % without causing significant overhydrogenation and then to carry out hydroformylation, methoxycarbonylation, and ethenolysis with significantly increased activity (up to twentyfold) and selectivity (up to 80 % increase). With these findings, we hope to convince the scientific and industrial world of the potential of selective partial hydrogenation as a key technology for utilizing renewable raw materials and to encourage its effective use in future work.