TU Dresden Data Publications

Data publications from research of Dresden University of Technology.

Browse

Recent Submissions

Now showing 1 - 5 of 61
  • Item
    Open Access
    Data for Simon Nogo_NFL_nCREANN
    (Technische Universität Dresden, 2025-03-13) Elmers, Julia Kristina; Mückschel, Moritz; Akgün, Katja; Ziemssen, Tjalf; Beste, Christian
    The data set includes raw behavioral data (logfiles) from the Simon Nogo task, raw preprocessed EEG data, and NFL data of 55 healthy participants. Further, all customized scripts for the analyses are provided.
  • Item
    Open Access
    OC_Identifier - Programm to use OC trained AI for the assessment of their state of maturation
    (Technische Universität Dresden, 2025-03-11) Lv, Guofan; Kruppke, Benjamin
    This programm is part of the project, that has successfully trained an AI that can classify and calculate four cell types. This AI is practical and has a bright future in terms of cell identification. YOLOv5m was used for this AI training. With a better computer configuration, YOLOv5x can be used for training to obtain an even better AI model. Ultralytics is the software company that developed YOLOv5. It released the latest YOLO series version, YOLOv8, in January 2023. This new framework can lead to better training results. In addition to known improvement methods, AI technology has greater prospects. YOLOv5 can be combined with other AI frameworks to predict cell growth trends. AI can also make suggestions about cell culture conditions (temperature, media, frequency of media exchange, etc.) based on growth trends. The AI framework can also be combined with other devices, for example, to control robotic arms and microscopes to automatically complete image acquisition. With the help of this AI, laboratory staff can be freed from the tedious task of counting cells and do much more.
  • Item
    Open Access
    Learning Gentle Grasping Using Vision, Sound, and Touch
    (Technische Universität Dresden, 2025-03-11) Nakahara, Ken; Calandra, Roberto
    This dataset contains 1,500 robotic grasps collected for the paper of Learning Gentle Grasping Using Vision, Sound, and Touch. Additionally, we provide a description of this dataset and Python scripts to visualize the data and process raw data into a training dataset for a PyTorch model. The robotics system used consists of a multi-fingered robotic hand (16-DoF, Allegro Hand v4.0), 7-DoF robotic arms (xArm7), DIGIT tactile sensors, an RGB-D camera (Intel RealSense D435i), and a commodity microphone. The target object is a toy that emits sound when grasped strongly.
  • Item
    Open Access
    Raw data of the paper "Brillouin spectroscopic investigation of corneal hydration and the impact of cross-linking therapy on water retention"
    (Technische Universität Dresden, 2025-03-06) Rix, Jan; Steuer, Svea
    Raw Brillouin, Raman and PS-OCT data acquired within the project and paper "Brillouin spectroscopic investigation of corneal hydration and the impact of cross-linking therapy on water retention"
  • Item
    Open Access
    Flossgraben Bridge - Large-Scale Experiment Acceleration Data
    (Technische Universität Dresden, 2025-03-05) Rohrer, Maximilian; Lenzen, Armin
    The research presented is part of the SPP100+ of the German Research Foundation. Subject of the research is the early detection of damage, especially on large bridge structures. For this purpose, a large-scale experiment for vibration-based output-only damage detection was conducted at Flossgraben Bridge near Zeitz (Germany). The experiment was conducted over the course of three days for which the structure was equipped with 56 uniaxial piezoelectric acceleration sensors. The experimental structure is a healthy bridge structure constructed in 2001 and is part of the federal road B2 with two lanes for traffic. The bridge is composed of seven spans, with a total length of 358 m and a total weight of approximately 4750 t. The superstructure consists of an in-situ concrete slab resting on a trapezoidal, torsionally rigid steel box girder with inclined webs. The large-scale experiment was divided into a reference phase and two damage equivalent load case phases. Cargo trucks with a total mass of 39 t were used as additional mass and placed in field 4 and field 3 as mass alterations. During the complete course of the large-scale experiment one traffic lane was closed, while the other was open for traffic. The experiment was carried out with the kind support of the LSBB Saxony-Anhalt.