Fraunhofer Institute for Wind Energy Systems
Fordatis - Forschungsdaten-Repositorium der Fraunhofer-GesellschaftNot a member yet
228 research outputs found
Sort by
FlexSense Infrastructure Sensor Dataset: Road Side Infrared Image and 4D Radar Point Cloud for Traffic Analysis
No full textDataset for Development of Algorithms for Object Detection and Tracking Using Sensor Data Fusion of Infrared Image and Radar Point Cloud in Traffic Analysis
INFRA-3DRC Dataset
No full textThe INFRA-3DRC Dataset is a public dataset comprised of 3D automotive Radar and RGB camera data generated using an intelligent roadside infrastructure (also known as smart infrastructure) setup.
This work is supported by the Bavarian Ministry of Economic Affairs, Regional Development and Energy (StMWi), Germany within the Project “InFra — Intelligent Infrastructure.
InVar-100: Industrial Objects in Varied Contexts Dataset
Get PDFThe Industrial Objects in Varied Contexts (InVar) dataset was internally produced by our team and contains 100 objects in 20800 total images (208 images per class). The objects consist of common automotive, machine and robotics lab parts. Each class contains 4 sub-categories (52 images each) with different attributes and visual complexities. White background (Dwh): The object is against a clean white background and the object is clear, centred and in focus. Stationary Setup (Dst): These images are also taken against a clean background using a stationary camera setup, with uncentered objects at a constant distance. The images have lower DPI resolution with occasional cropping. Handheld (Dha): These images are taken with the user holding the objects, with occasional occluding. Cluttered background (Dcl): These images are taken with the object placed along with other objects from the lab in the background and with no occlusion. The dataset was produced to simulate the miscellaneous issues in industrial setups as discussed. The dataset was produced by our staff at different workstations and labs in Berlin. More details regarding the objects used for digitisation are available in the metadata file
Quellcode zur Simulation im Projekt (BGA-PtG)2: Ganzheitliche Bewertung der Integration von Power-to-Gas-Konzepten in Biogas- und Biomethananlagen einschließlich der Entwicklung von Geschäftsmodellen für regenerative Gase
No full textQuellcode einer technischen und wirtschaftlichen Simulation. Kontext ist das Forschungsprojekt (BGA-PtG)2: Ganzheitliche Bewertung der Integration von Power-to-Gas-Konzepten in Biogas- und Biomethananlagen einschließlich der Entwicklung von Geschäftsmodellen für regenerative Gase.
Der Abschlussbericht wird bei der TIB veröffentlicht
Spatial analysis of renewable and excess heat potentials for climate-neutral district heating in Europe
No full textSupplementary datasets to the journal paper in excel format for results and GIS-datasets (raster and vector) for dataset
Hybrid quantum transfer learning for crack image classification on NISQ hardware
No full textThis source code is meant to support the understanding of our paper Hybrid quantum transfer learning for crack image classification on NISQ hardware. Quantum computers possess the potential to process data using a remarkably reduced number of qubits compared to conventional bits, as per theoretical foundations. However, recent experiments have indicated that the practical feasibility of retrieving an image from its quantum encoded version is currently limited to very small image sizes. Despite this constraint, variational quantum machine learning algorithms can still be employed in the current noisy intermediate scale quantum (NISQ) era. An example is a hybrid quantum machine learning approach for edge detection. In our study, we present an application of quantum transfer learning for detecting cracks in gray value images. We compare the performance and training time of PennyLane’s standard qubits with IBM’s qasm_simulator and real backends, offering insights into their execution efficiency.This work was supported by the project AnQuC-3 of the Competence Center Quantum Computing Rhineland-Palatinate (Germany) and by the German Federal Ministry of Education and Research (BMBF) under grant 05M2020 (DAnoBi). Additionally, this work was funded by the Federal Ministry for Economic Affairs and Climate Action (German: Bundesministerium für Wirtschaft und Klimaschutz) under the project EniQmA with funding number 01MQ22007A.Mainly used python, together with qiskit, pennylane and torch. Specific versions of the packages are written in the jupyter notebook or in the requirements.txt or environment.yml. See there for more informatio
iPSC-Kolonie, aufgenommen mit High-Speed-Mikroskop
No full textImages of iPSCs taken in the StemCellFactory (more information about the project: Elanzew A, Nießing B, Langendoerfer D et al. (2020) The StemCellFactory: A Modular System Integration for Automated Generation and Expansion of Human Induced Pluripotent Stem Cells. Front. Bioeng. Biotechnol. 8:580352. doi: 10.3389/fbioe.2020.580352). The images show induced pluripotent stem cells at different times during their development. A Nikon Ti-Eclipse phase contrast microscope equipped with faster stage (Märzhäuser Wetzlar Scan IM 130 x 85), camera (PCO Edge 4.2) and lighting was used for all images.Bilder von iPS, die in der StemCellFactory aufgenommen wurden (weitere Informationen über das Projekt: Elanzew A, Nießing B, Langendoerfer D et al. (2020) The StemCellFactory: A Modular System Integration for Automated Generation and Expansion of Human Induced Pluripotent Stem Cells. Front. Bioeng. Biotechnol. 8:580352. doi: 10.3389/fbioe.2020.580352). Die Bilder zeigen induzierte pluripotente Stammzellen zu verschiedenen Zeitpunkten ihrer Entwicklung. Für alle Untersuchungen wurde ein Nikon Ti-Eclipse Phasenkontrastmikroskop mit schnellerem Tisch (Märzhäuser Wetzlar Scan IM 130 x 85), Kamera (PCO Edge 4.2) und Beleuchtung verwendet.grant no. 005-1007-0021 (StemCellFactory I)grant no. 005-1403-0102 (StemCellFactory II)grant no. EFRE-0800978 (StemCellFactory III)Ministerium für Wirtschaft, Industrie, Klimaschutz und Energie NR
ÖFIT Bevölkerungsumfrage 2019
No full textIn dieser Umfrage geht es um die Haltung der Befragen zur (digitalen) öffentlichen Verwaltung und dabei insbesondere um das „Once Only“-Prinzip, also die Idee, dass Bürger:innen ihre Daten nur einmalig an die Verwaltung übermitteln und diese dann zwischen unterschiedlichen Akteuren ausgetauscht werden. Die Fragen drehen sich um die Akzeptanz von Once Only, die Freigabe von Informationen an Unternehmen sowie die Wünsche bezüglich Transparenz und Kontrolle über den Datenfluss. Es werden Unterschiede in der Akzeptanz hinsichtlich verschiedener Arten persönlicher Daten, bspw. Finanzdaten oder Gesundheitsdaten, sowie hinsichtlich verschiedener Szenarien, bspw. Umzüge oder Geburten, abgefragt.
Für die Umfrage wird neben dem Datensatz (in zwei Versionen: sowohl mit Werten als auch mit Label) auch der Fragebogen mit dem Codeschema bereitgestellt. Zusätzlich werden die vom Dienstleister verfügbaren Angaben zu den Methoden der Datenvalidierung, Bereinigung und Gewichtung der Datensätze bereitgestellt.Die Daten werden jeweils in zwei Formaten zugänglich gemacht. Formate für Daten (mit Werten und mit Label): .csv und .xlsx Formate für Fragebogen: .docx und .odt Formate für Methodenberichte: .pd
InVar-100: Industrial Objects in Varied Contexts Dataset
Get PDFThe Industrial Objects in Varied Contexts (InVar) dataset was internally produced by our team and contains 100 objects in 20800 total images (208 images per class). The objects consist of common automotive, machine and robotics lab parts. Each class contains 4 sub-categories (52 images each) with different attributes and visual complexities. White background (Dwh): The object is against a clean white background and the object is clear, centred and in focus. Stationary Setup (Dst): These images are also taken against a clean background using a stationary camera setup, with uncentered objects at a constant distance. The images have lower DPI resolution with occasional cropping. Handheld (Dha): These images are taken with the user holding the objects, with occasional occluding. Cluttered background (Dcl): These images are taken with the object placed along with other objects from the lab in the background and with no occlusion. The dataset was produced to simulate the miscellaneous issues in industrial setups as discussed. The dataset was produced by our staff at different workstations and labs in Berlin. More details regarding the objects used for digitisation are available in the metadata file
CARLA Scenarios
Get PDFResults for autonomous driving scenarios simulated with CARLA under different environment conditions.This work was supported by the Bavarian Ministry of Economic Affairs, Regional Development and Energy through the Center for Analytics–Data–Applications (ADACenter) within the framework of ”BAYERN DIGITAL II”.The models have been trained on the datasets Berkely Deep Drive, COCO and KITTI with ensemble distribution distillation and standard softmax training. As base model architecture a slightly advanced version of Yolov3 has been used