TU Wien Research Data
Not a member yet
512 research outputs found
Sort by
Reality Capture automated image alignment
No full text<p>This video shows the use of Reality Capture photogrammetric structure-from-motions software in combination with a bat-script for automated processing of images into a 3D model. It was screen-recorded during testing as part of an automated building documentation workflow for architectural heritage. </p><p>The bat-script is used to monitor a certain folder in Windows for incoming images - transferred from a camera or tabletPC to the computer. After a user-defined amount of images is in the folder, Reality Capture will be initiated to align images and export the resulting registration in csv format. As the parameters for registration export are set to geodetic coordinates, the script is intended to be used with images with a GNSS position within EXIF metadata. </p><p>Within the video, an adapted version from RealityCapture CLI Sample Script "Repetitively check for new images in the image folder and align them with already imported images" is used. </p><p><a href="https://www.capturingreality.com/realitycapture-cli-sample-scripts">https://www.capturingreality.com/realitycapture-cli-sample-scripts</a> (last accessed April 17th, 2024)</p><p>The video of the script in action can also be found here: <a href="https://youtu.be/yz2PH3t7CGQ?si=9EtUCr_0ieif1iZB">https://youtu.be/yz2PH3t7CGQ?si=9EtUCr_0ieif1iZB</a></p><p>(top left batch script, top right Reality Capture instance, bottom left input data, bottom right export data)</p>
Mobility Market Indicators and Macroeconomic Indicators for the MaaS Status Index (MSI) in Austria (2017-2028)
No full text<h2>Dataset description</h2><p>The datasets include mobility market indicators and macroeconomic indicators for Austria, which were used to calculate the Mobility as a Service (MaaS) Status Index (MSI). The MSI evaluates the readiness and potential for implementing Mobility as a Service (MaaS) in Austria. The datasets cover two distinct periods: 2017-2022 (T1) and 2023-2028 (T2). The indicators include annual revenues, vehicle costs, number of users, market shares, GDP per capita, urbanization rates, and investments in transportation infrastructure, among others.</p><h3>Context and methodology</h3><p>Each indicator is represented by the average annual growth rate, a mean value, and a normalized mean value (min-max-normalization) for period T1 and T2. The data were sourced from <a href="https://www.statista.com/outlook/mobility-markets ">Statista (2024)</a></p><h3>Technical details</h3><p>The dataset contains two Microsoft Excel files (one for mobility market indicators, one for macroeconomic indicators). Other than Microsoft Excel, there is no additional software needed to investigate the data.</p><h3> </h3>
Nanditha Kattukudiyil Narayanan: 2-(o-Tolyl) Pyridine as Ligand Improves the Efficiency in Ketone Directed ortho-Arylation
No full text<p>Uploaded files:</p>
<p>1. Zip-folder: NMR data files (fid) for all newly synthesized compounds of this project. The fids can be opened with all standard NMR-evaluation software (e.g. MestreNova, TopSpin, ACD-Labs, etc.)</p>
<p>2. pdf document with a conversion table of the publication numbers of compounds into the number from the measured files</p>
<p>3. pdf document with the supporting info for the publication "2-(o-Tolyl) Pyridine as Ligand Improves the Efficiency in Ketone Directed ortho-Arylation"</p>
<p> </p>
Data for: "Waterfalls: umbilical cords at the birth of Hubbard bands"
No full text<h3>Context and methodology</h3>
<ul>
<li>This repository contains raw data from the associated research work. It serves the purpose of aiding interested readers to reproduce the results of the related work and verify their validity.</li>
<li>The research area in which this dataset is created is that of condensed matter, strongly correlated electron systems, ARPES, cuprates, and nickelates.</li>
<li>This dataset was created using the <a href="https://github.com/w2dynamics/w2dynamics">w2dynamics</a> code for DMFT calculations, the <a href="https://github.com/josefkaufmann/ana_cont">ana_cont</a> package for analytic continuation, and the <a href="https://github.com/PaulWorm/DGApy">DGApy</a> code for DGA calculations. </li>
</ul>
<h3>Technical details</h3>
<ul>
<li>For each figure there is a corresponding folder in data folder containing the data shown in the figure. Data files are of type hk, hdf5, npz, and txt and named logically after data they contain.</li>
<li>Reading the data requires python (version 3.10.13, numpy version 1.26.0) with h5py (version 3.10.0).</li>
<li>The detailed content of the data files can found in a README file.</li>
</ul>
<h3>w2dynamics calculations</h3>
<ul>
<li>DMFT calculations are done in three iterations, each involving different numbers of DMFT steps and statistics.</li>
<li>Corresponding input files _Parameters_dmft_iter.in, _Parameters_dmft_iter_2.in, and _Parameters_dmft_stat.in can be found in the folder input_files/w2dynamics/, and are executed in that order.</li>
<li>The output of the last iteration is contained in 1p-data.hdf5, including the self-energy on the Matsubara axis.</li>
</ul>
<h3>DGApy calculations</h3>
<ul>
<li> In addition to the single-particle data in 1p-data.hdf5, DGApy requires the two-particle DMFT data contained in g4iw_sym.hdf5.</li>
<li>The corresponding input file _Parameters_vertex.in for w2dynamics can be found in input_files/w2dynamics/. Note that the DMFT output will contain the full Vertex.hdf5. To get g4iw_sym.hdf5 one needs to execute sym1b (part of DGApy) in the same folder where Vertex.hdf5 is located.</li>
<li>The resulting self-energy on the Matsubara axis is contained in siwk_dga.npy files.</li>
</ul>
<h3>ana_cont</h3>
<ul>
<li>Analytical continuation of the self-energy to the real-frequency axis was performed using MaxEnt as implemented in the ana_cont python package; model for fitting was flat and chi2kink method was used to determine the hyperparameter alpha.</li>
<li>For temperatures T = 100/t and T = 15/t, number of fermionic Matsubara frequencies taken was 800 and 200, while the real-frequency grid was taken to be linear with 1001 points in the range [-50,50].</li>
<li>Errors for DMFT self-energies can be found in 1p-data.hdf5, while for DGA results it is taken to be 1e-3; preblur was set to 1e-3 in all calculations.</li>
<li>Analyically continued DMFT self-energy is located in self_energy.npz files, while analytically continued DGA self-energy in self_energy_nw1001_nk400_delta0.04.npy. Note that the interpolation in both momentum and frequency space may be employed to get better resolution. In addition, a constant imaginary part of 0.04 was added to the DGA self-energy to improve stability of results.</li>
</ul>
Dataset for Vertical Jump Height Estimation from Depth Camera and Wearable Accelerometer Motion Data
No full text<h2>Dataset for Vertical Jump Height Estimation from Depth Camera and Wearable Accelerometer Motion Data</h2>
<p>While the training of vertical jumps offers benefits for agility and performance across various amateur sports, the objective measurement of jump height remains a challenge compared to simpler assessments like the broad jump distance in a sand pit. Aiming at the <strong>estimation of the vertical jump height</strong> with easy-to-use and cost-efficient devices, we recorded a comprehensive dataset with an <strong>off-the-shelf depth camera</strong> and <strong>cost-efficient wearable motion sensors</strong>, equipped with an onboard three-axis accelerometer sensor. In our <strong><a title="JumpMetric: Assessment of Fiducial Positions for Vertical Jump Height Estimation from Depth Cameras and Wearable Sensors" href="https://doi.org/10.1145/3701571.3701607" target="_blank" rel="noopener">publication</a></strong>, we assessed the accuracy achievable at diverse fiducial positions, which are <strong>7 skeletal joints</strong> from the depth camera and <strong>10 wearing positions</strong> of the sensing devices. The <strong>user study</strong> was conducted with <strong>44 subjects</strong> (33 male, 11 female, 23.1 ± 2.2 years) performing <strong>five countermovement jumps</strong> each. In order to gather <strong>ground truth</strong> information, a conventional digital camera was used to document the jumps and the vertical hip displacement along a measuring tape. Thales’ theorem on proportionality was then applied to <strong>rectify the perspective displacement</strong> of the manual readings from the video footage.</p>
<h3>Context and Methodology</h3>
<ul>
<li>Dataset for research on the estimation of <strong>vertical jump height</strong> and in adjacent fields such as <strong>human activity recognition</strong> etc.</li>
<li>The dataset provides recordings from two easy-to-use and cost-efficient sensing modalities, the off-the-shelf <strong>depth camera</strong> Microsoft Azure Kinect and <strong>10 wearable 3-axis accelerometers</strong></li>
<li>The <strong>ground truth</strong> information are manually determined and rectified</li>
<li>The dataset was used in a <strong><a title="JumpMetric: Assessment of Fiducial Positions for Vertical Jump Height Estimation from Depth Cameras and Wearable Sensors" href="https://doi.org/10.1145/3701571.3701607" target="_blank" rel="noopener">publication</a></strong> showing that the most accurate estimates of the depth camera were obtained from the pelvis and thoracic spine joint (error of -15.8 ± 23.3mm and 24.2 ± 35.1 mm) while the best estimates from the wearable motion sensor data were obtained from the neck position and the ankles (error of 18.8 ± 29.0mm and -4.8 ± 35.2 mm)</li>
<li>We <strong>encourage researchers</strong> to improve on our findings, e.g., by applying advanced <strong>machine learning</strong> techniques on the provided dataset</li>
</ul>
<h3>Technical Details</h3>
<ul>
<li>A total of <strong>220 recordings</strong> of <strong>countermovement jumps</strong>: 44 subjects, 5 jumps, two sensing modalities, manually determined and rectified ground truth</li>
<li><strong>44 subjects</strong>: 33 male and 11 female with an average age of 23.1 ± 2.2 years</li>
<li>Subjects gave <strong>written consent</strong> to provide the measurements for research purposes and publication (video recordings were deleted after ground truth determination)</li>
<li><strong>Depth camera</strong>: Microsoft Azure Kinect, 3d-coordinates (x, y, and z) for all <strong>32 joint positions</strong> recorded along with the accompanied timestamp, frame rate of <strong>30 Hz</strong></li>
<li><strong>Wearable motion sensors</strong>: 10 wearing positions (lower neck, chest (sternum), hips (left and right), thighs (left and right), ankles (left and right), and wrists (left and right)), 3-axis accelerometer sensor data (x, y, and z) along with a timestamp for each sample, sampling rate of <strong>100 Hz</strong></li>
<li>Two folders 'depthcamera' and 'wearables', each containing 44 subfolders labeled with the subject ids '01' to '44'</li>
<li>Every subject's folder again contains 5 subfolders labeled with the jump ids 'j1' to 'j5' that contain the files of the countermovement jump recordings</li>
<li>The recordings are provided in both Python pickle files *.p as well as comma-separated value *.csv (';' as separator) files with the file names composed of subject id and jump id for *.p files, e.g., 's05_j3.p', as well as the joint or wearing position for the *.csv files, e.g., 's05_j3_hip_left.csv'</li>
<li>The pickle files have been tested successfully with Python 3.13.1 and NumPy 2.2.2</li>
<li>Unfortunately, for <strong>subject 02</strong>, the accelerometer data of neck and chest were not successfully recorded and, hence, the associated lists in the *.p and the *.csv files empty</li>
<li><strong>Demographic information</strong>: a summary of the individual subjects' demographic information is available in the files 'subjects.p' or 'subjects.csv', providing the gender as female or male, age in years, height in cm, weight in kg, if they were conveyed by the lecture, if they were a student and, if so, of which degree</li>
<li><strong>Ground truth</strong>: the manually determined and rectified ground truth information are provided in the files 'groundtruth.p' or 'groundtruth.csv', associated with the individual subject ids, providing jump number, jump height, and whether the jump is considered an overall well-executed jump</li>
<li>The <strong>dataset description</strong> at hand is also provided in the 'README.txt' file of the dataset's *.zip file</li>
</ul>
Numerical results for "Protection of correlation-induced phase instabilities by exceptional susceptibilities"
No full text<p>This data repository contains the original figures, numerical (raw) data, and plot scripts to reproduce the figures from the publication "Protection of Correlation-Induced Phase Instabilities by Exceptional Susceptibilities" at <a href="https://doi.org/10.1103/PhysRevResearch.6.L022031">Physical Review Research</a>. LaTeX source files of the preprint available on <a href="https://doi.org/10.48550/arXiv.2307.00849">arXiv</a> can be found at the TU <a href="https://gitlab.tuwien.ac.at/e138/e138-02/protection-of-correlation-induced-phase-instabilities-by-exceptional-susceptibilities">gitlab</a> repository. Additional information can be found in the README.</p><h3><strong>License</strong></h3><p>The CC-BY license applies to all the data and pdf files. All distributed code is under the MIT license.</p>
Data for: "Sputter Yields of the Lunar Surface: Experimental Validation and Numerical Modelling of Solar Wind Sputtering of Apollo 16 Soils"
No full text<h2>Introduction</h2>
<p>This data repository contains the data behind the figures of the manuscript entitled "Sputter Yields of the Lunar Surface: Experimental Validation and Numerical Modelling of Solar Wind Sputtering of Apollo 16 Soils". The CC-BY licence applies to all data files. </p>
<h2>Figure 2</h2>
<p>Data files behind Figure 2 (sputter yields for flat samples) have names starting with "Fig2a" and "Fig2b", indicating data for hydrogen and helium irradiations, respectively. The rest of the filename coincides with the labelling of the Figure.<br>They are organised as comma separated value (CSV) files, where the first line gives the names and units of the columns. For this Figure, data are given in both units amu/ion and atoms/ion, corresponding to both axes of the Figure. </p>
<h2>Figure 3 and Ejecta Angular Distributions</h2>
<p>Ejecta angular distributions as measured by the catche QCM technique are given by the files starting with "Angular_emission", where the incidence angle, the projectile species and the sample type are denoted by the rest of the filename. Data are organised as CSV files where the names of the columns are given in the first lines. Angles are measured in degrees, catcher mass yields (mass deposited onto the catcher QCM normalised per incoming ion and detector solid angle) as well as the absolute errors are given amu/ion.</p>
<p>Data are made available for H and He under both 60° and 45° incidence for both sample types in the repository. In the manuscript, only 60° incidence of 4 keV He is shown for illustratory purposes. To recreate this Figure 3, the data are stored in "Angular_emission_60deg_He_rough.dat" and "Angular_emission_60deg_He_flat.dat".</p>
<p>Note that for the hydrogen case, data are given normalised per H2 ion.</p>
<h2>Figure 4</h2>
<p>Data that occurs for the first time in Figure 4 is stored in files starting with "Fig4a" and "Fig4b", respectively. The rest of the naming convention follows the labelling from the Figure legend, allowing to uniquely identify the data sets. Some data occur first in Figure 1. We refrained from uploading these data sets a second time. </p>
<h2>Figure 5</h2>
<p>No new original data are presented in Figure 5. Data that was created and analysed for this study was already shown in the previous Figures. Data from other publications are clearly identified in the text. We do not upload these data from different publications in this repository. Instead, the interested reader is referred to the original publications.</p>
Supplemental material of the Diploma Thesis "Holotomographic microscopy of subcellular structures and nuclear dynamics in filamentous fungi"
No full text<p>The supplemental online material of the Diploma Thesis "Holotomographic microscopy of subcellular structures and nuclear dynamics in filamentous fungi" includes two videos and an SOP (standard operating procedure) for data processing. Each file is briefly described below. For a more detailed description of the events depicted in the videos, please refer to the Diploma Thesis.</p><p> </p><p>Video 1 shows the nuclei movement of the engineered Aspergillus niger strain A621_H2A:sGFP. The strain produces the histone H2A tagged with sGFP under the control of the inducible tet-on promoter. Images were recorded with the 3D Cell Explorer-fluo (Nanolive SA, Tolochenaz, Switzerland). The recording visualizes the movement of nuclei by superimposing the fluorescence signal onto the maximum intensity projection of the refractive index map. </p><p>Observation of synchronous nuclei division is possible in the apical hyphal compartments located on the left and right sides of the focused hypha. In the middle, a single nucleus is located in a subapical compartment separated by two septa. The nucleus does not undergo any significant movement or division during the observed time frame and is therefore considered mitotically inactive.</p><p> </p><p>Video 2 shows the behavior of the cyclin-dependent kinase nimX in the engineered Aspergillus niger strain A621_nimX:sGFP. The strain produces GFP-tagged nimX under the control of the inducible tet-on promoter. Images were recorded with the 3D Cell Explorer-fluo. The video displays the maximum intensity projection of the holotomographic slices on the left in black and white. On the right, the signal of the GFP-tagged nsimX is shown in green.</p><p>The video demonstrates the emergence of two germ tubes and the movement of accumulated nimX towards the growing hyphal tip. At this point, nimX is likely located in the nuclei, as previously reported for Aspergillus nidulans. In the last frame of the video, the GFP signal suddenly disappears, indicating that nimX is either exiting the nuclei or undergoing degradation.</p><p> </p><p>The SOP aims to assist with processing data from the 3D Cell Explorer-fluo, mainly in Fiji/ImageJ. A short part describes data acquisition and export with the microscope software Steve. The fungi Aspergillus niger, Trichoderma reesei and Aureobasidium pullulans were used to create this SOP, so there may be differences in data processing when working with other organisms.</p><p> </p>
Road Network Mapping from Multispectral Satellite Imagery: Leveraging Deep Learning and Spectral Bands
No full text<h2>Road Network Mapping from Multispectral Satellite Imagery: Leveraging Deep Learning and Spectral Bands</h2><p>Submitted to AGILE24</p><h3>Abstract</h3><p>Updating road networks in rapidly changing urban landscapes is an important but difficult task, often challenged by the complexity and errors of manual mapping processes. Traditional methods that primarily use RGB satellite imagery struggle with obstacles in the environment and varying road structures, leading to limitations in global data processing. This paper presents an innovative approach that utilizes deep learning and multispectral satellite imagery to improve road network extraction and mapping. By exploring U-Net models with DenseNet backbones and integrating different spectral bands we apply semantic segmentation and extensive post-processing techniques to create georeferenced road networks. We trained two identical models to evaluate the impact of using images created from specially selected multispectral bands rather than conventional RGB images. Our experiments demonstrate the positive impact of using multispectral bands, by improving the results of the metrics Intersection over Union (IoU) by 6.5%, F1 by 5.4%, and the newly proposed relative graph edit distance (relGED) and topology metrics by 2.2% and 2.6% respectively.</p><h3>Data</h3><p>To use the code in this repository, download the required data from SpaceNet Challenge 3 (https://spacenet.ai/spacenet-roads-dataset/) via AWS.</p><p>The SpaceNet Dataset by SpaceNet Partners is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.</p><p>SpaceNet was accessed on 05.01.2023 from https://registry.opendata.aws/spacenet</p><h3>Software</h3><p>The analysis and results of this research were achieved with Python and several software packages such as:</p><p>- tensorflow</p><p>- networkx</p><p>- Pillow, cv2</p><p>- GDAL, rasterio, shapely</p><p>- APLS</p><p>For a fully reproducible environment and software versions refer to 'environment.yml'.</p><p>All data is licensed under CC BY 4.0, all software files are licensed under the MIT License.</p><h3>Reproducibility</h3><p>To execute the scripts and train your model, first refer to the 'Data' section of this file to download the data from the providers. Apply the preprocessing steps from 'preprocessing.py', but consider that to avoid redundancy, preprocessing steps not included in this repository are the conversion of geojson road data into training images, the reduction of satellite images to an 8-bit format, and their conversion into '.png' files. These steps can be achieved by applying and, if necessary, modifying the APLS library which is publicly available under https://github.com/CosmiQ/apls. Apply preprocessing to both RGB and MS images. To generate the latter execute the 'ms_channel_seperation.py' script while specifying the wanted multispectral channels. Execute the 'train_model.py' script to train your semantic segmentation model, and apply post-processing procedures with 'postprocessing.py'. Generate the metrics results by executing 'evaluation.py'.</p><p>To save storage space, not all the used data is made available in this repository. Please refer to the 'Data' section of this file to access and download the data from the providers. Exemplary preprocessed training data (100 split images of Las Vegas) is included in the folders './data/tiled512/small_test_sample/ms/' and './data/tiled512/small_test_sample/rgb/'. Post-processed results are provided in the corresponding folders './results/UNetDense_MS_512/' and './results/UNetDense_RGB_512/'. These include the stitched and recombined images, without any post-processing applied to them, as well as the extracted and post-processed graphs as '.pickle' files. This provided data was used to calculate the metrics Intersection over Union (IoU), F1 score, relGED, and topology metric as presented in the paper.</p><p>The figures included in the paper can be reproduced by saving images created during the preprocessing, training, and post-processing steps. To generate the plots of resulting graphs, refer to the corresponding functions and enable the boolean parameter 'plot'. Bounding boxes seen in the figures were drawn manually and only serve an explanatory purpose.</p><p>Please be advised that file paths and folder structure have to be adapted manually in the scripts to suit the users folder structure. Be aware of selecting uniform file paths and storing the results in folders named after their model. Furthermore, the code is not meant to be executed from the terminal, running the individual scripts in an IDE is recommended.</p>
Chemo-mechanical Toolbox
No full text<p><span>With this document, we aim to provide a brief, yet comprehensive overview of standardized, as well as advanced methods to analyze bituminous materials. For each method, the general principles are laid out and the specific application of bitumen is explained. References to publications related to the methods are also included. To make the document more useful for practitioners, we provide a short summary of potential applications and sample questions that can be answered by a method, its advantages, disadvantages, costs, and limitations. </span></p>
<p><span>This document, which we call Chemo-Mechanical Toolbox, intends to provide short pieces of information for the application of methods. It is not intended to be a textbook on the fundamental background of analytical principles, or a review of findings made by applying these methods in the past. We want to provide a reference document that supports users finding the right method for different questions on bitumen’s chemistry, mechanics, microstructure, and aging – and how all these phenomena are linked. </span></p>