@techreport{SparrerWinkelmannvonHoffmannetal.2023, type = {Working Paper}, author = {Sparrer, Lisa Marie and Winkelmann, Sven and von Hoffmann, Alexander and S{\´a}nchez Morillo, Daniel}, title = {A Novel Framework for the Assessment of the Impact of Affective UI Sounds in a Driving Situation Through Virtual Reality and Physiological Signals}, doi = {10.34646/thn/ohmdok-1202}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:92-opus4-12021}, pages = {1 -- 7}, year = {2023}, abstract = {The objective of this study was to develop an experimental framework that allows research on whether the combined use of a virtual reality (VR) driving simulator, electroencephalography signals collected during driving tests and designed attentional paradigms could be applied to assess the performance of affective warning sounds for re-channelling driver's attention in certain driving situations. Three VR driving scenarios with different stimuli and two driving tasks were designed and implemented. Acquisition of driver's physiological data and behavioural metrics were integrated to be gathered synchronously. In addition, an experimental protocol was designed for the assessment of the impact of affective warning sounds. The technical feasibility of the framework was evaluated with the results showing that the proposed framework enables the data acquisition and analysis for a variety of warning stimuli. After successful technical deployment, a study with experimental subjects will follow to evaluate the effectiveness of affective warning sounds.}, language = {en} } @inproceedings{RennerGleichaufWinkelmann2024, author = {Renner, Philipp and Gleichauf, Johanna and Winkelmann, Sven}, title = {Non-Contact In-Car Monitoring of Heart Rate: Evaluating the Eulerian Video Magnification Algorithm in a Driving Simulator Study}, series = {Proceedings of Mensch und Computer 2024}, booktitle = {Proceedings of Mensch und Computer 2024}, publisher = {ACM}, address = {New York, NY, USA}, doi = {10.1145/3670653.3677493}, pages = {651 -- 654}, year = {2024}, abstract = {Monitoring drivers' health is crucial for saving lives in emergencies and enabling in-car health applications. The state of the art in pulse monitoring is contact-based sensors which impair the driving experience and have to be applied manually before driving. This paper focuses on automated hyper parameter optimizing the Eulerian Video Magnification (EVM) algorithm, which detects heart rates through non-contact facial camera images, for use in driving scenarios. We conducted a user study where 21 participants performed a driving simulation while their heart rates were recorded by a wearable fitness tracker (serving as ground truth) and facial images with an RGB camera. Our findings indicate that, despite using the optuna library for hyper parameter tuning, the Eulerian Video Magnification algorithm is insufficient for accurate pulse detection in a driving simulator environment.}, language = {en} } @inproceedings{WinkelmannBuettnerDeivasihamanietal.2023, author = {Winkelmann, Sven and B{\"u}ttner, Max and Deivasihamani,, Dharani and von Hoffmann, Alexander and Flohr, Fabian}, title = {Using Node-RED as a Low-Code Approach to Model Interaction Logic of Machine-Learning-Supported eHMIs for the Virtual Driving Simulator Carla}, series = {AutomotiveUI '23 Adjunct: Adjunct Proceedings of the 15th International Conference on Automotive User Interfaces and Interactive Vehicular Applications}, booktitle = {AutomotiveUI '23 Adjunct: Adjunct Proceedings of the 15th International Conference on Automotive User Interfaces and Interactive Vehicular Applications}, publisher = {ACM}, doi = {10.1145/3581961.3609844}, pages = {323 -- 326}, year = {2023}, abstract = {External Human-Machine Interfaces (eHMI) enable interaction between vehicles and Vulnerable Road Users (VRU), for example, to warn VRUs of the car's presence. Warning systems should warn of the situation's urgency, which can be achieved using Machine Learning (ML)-based VRU detection models. ML models and eHMI interaction concepts are usually developed by different teams and tested separately, often resulting in integration problems. This work contributes to a low-code approach to model interaction concepts involving ML models to enable end-to-end prototypes for early integration and User eXperience (UX) testing. We use flow-based modeling with Node-RED, the virtual driving simulator CARLA and YOLOv5 as state-of-the-art deep learning techniques for VRU detection. We show two scenarios (cornering lights and context-aware VRU warning) in an interactive demonstrator, meaning a manual live control of pedestrian and car. We consider our approach to model and evaluate interaction concepts without writing code feasible for non-computer scientists.}, language = {en} } @incollection{DalferthWinkelmannSchwenker2020, author = {Dalferth, Jasmin and Winkelmann, Sven and Schwenker, Friedhelm}, title = {Using Mask R-CNN for Image-Based Wear Classification of Solid Carbide Milling and Drilling Tools}, series = {Artificial Neural Networks in Pattern Recognition}, booktitle = {Artificial Neural Networks in Pattern Recognition}, publisher = {Springer International Publishing}, address = {Cham}, isbn = {9783030583088}, issn = {0302-9743}, doi = {10.1007/978-3-030-58309-5_18}, pages = {223 -- 234}, year = {2020}, abstract = {In order to ensure high productivity and quality in industrial production, early identification of tool wear is needed. Within the context of Industry 4.0, we integrate wear monitoring of solid carbide milling and drilling cutters automatically into the production process. Therefore, we propose to analyze wear types with image instance segmentation using Mask R-CNN with feature pyramid and bounding box regression. Our approach is able to recognize the five most important wear types: flank wear, crater wear, fracture, built-up edge and plastic deformation. While other methods use image classification and classify only one wear type for each image, our model is able to detect multiple wear types. Over 35 models with different hyperparameter settings were trained on 5,000 labeled images to establish a reliable classifier. The results show up to 82.03\% accuracy and benefit for overlapping wear types, which is crucial for using the model in production.}, language = {en} } @article{TomaBoenischLoehnhardtetal.2024, author = {Toma, Monica and B{\"o}nisch, Caroline and L{\"o}hnhardt, Benjamin and Kelm, Michael and Bohnenberger, Hanibal and Winkelmann, Sven and Str{\"o}bel, Philipp and Keszty{\"u}s, Tibor}, title = {Research collaboration data platform ensuring general data protection}, series = {Nature Scientific Reports}, volume = {14}, journal = {Nature Scientific Reports}, number = {1}, publisher = {Springer Science and Business Media LLC}, issn = {2045-2322}, doi = {10.1038/s41598-024-61912-8}, year = {2024}, abstract = {Translational data is of paramount importance for medical research and clinical innovation. It has the potential to benefit individuals and organizations, however, the protection of personal data must be guaranteed. Collecting diverse omics data and electronic health records (EHR), re-using the minimized data, as well as providing a reliable data transfer between different institutions are mandatory steps for the development of the promising field of big data and artificial intelligence in medical research. This is made possible within the proposed data platform in this research project. The established data platform enables the collaboration between public and commercial organizations by data transfer from various clinical systems into a cloud for supporting multi-site research while ensuring compliant data governance.}, language = {en} } @incollection{WinkelmannvonHoffmann2025, author = {Winkelmann, Sven and von Hoffmann, Alexander}, title = {Automotive Human-Machine-Interaction}, series = {Artificial Intelligence in Business and Engineering}, booktitle = {Artificial Intelligence in Business and Engineering}, editor = {Hofbauer, G{\"u}nter}, publisher = {Kohlhammer}, isbn = {978-3-17-046743-9}, doi = {10.17433/978-3-17-046743-9}, pages = {112 -- 128}, year = {2025}, language = {en} }