Publikationen an der Fakultät für Informatik und Automatisierung ab 2015

Anzahl der Treffer: 1905
Erstellt: Wed, 27 Mar 2024 23:30:23 +0100 in 0.0530 sec


Heijs, Janne J. A.; Havelaar, Ruben Jan; Fiedler, Patrique; Wezel, Richard J. A.; Heida, Tjitske
Validation of soft multipin dry EEG electrodes. - In: Sensors, ISSN 1424-8220, Bd. 21 (2021), 20, 6827, insges. 18 S.

Current developments towards multipin, dry electrodes in electroencephalography (EEG) are promising for applications in non-laboratory environments. Dry electrodes do not require the application of conductive gel, which mostly confines the use of gel EEG systems to the laboratory environment. The aim of this study is to validate soft, multipin, dry EEG electrodes by comparing their performance to conventional gel EEG electrodes. Fifteen healthy volunteers performed three tasks, with a 32-channel gel EEG system and a 32-channel dry EEG system: the 40 Hz Auditory Steady-State Response (ASSR), the checkerboard paradigm, and an eyes open/closed task. Within-subject analyses were performed to compare the signal quality in the time, frequency, and spatial domains. The results showed strong similarities between the two systems in the time and frequency domains, with strong correlations of the visual (p = 0.89) and auditory evoked potential (p = 0.81), and moderate to strong correlations for the alpha band during eye closure (p = 0.81-0.86) and the 40 Hz-ASSR power (p = 0.66-0.72), respectively. However, delta and theta band power was significantly increased, and the signal-to-noise ratio was significantly decreased for the dry EEG system. Topographical distributions were comparable for both systems. Moreover, the application time of the dry EEG system was significantly shorter (8 min). It can be concluded that the soft, multipin dry EEG system can be used in brain activity research with similar accuracy as conventional gel electrodes.



https://doi.org/10.3390/s21206827
Numssen, Ole; Zier, Anna-Leah; Thielscher, Axel; Hartwigsen, Gesa; Knösche, Thomas R.; Weise, Konstantin
Efficient high-resolution TMS mapping of the human motor cortex by nonlinear regression. - In: NeuroImage, ISSN 1095-9572, Bd. 245 (2021), 118654, insges. 11 S.

Transcranial magnetic stimulation (TMS) is a powerful tool to investigate causal structure-function relationships in the human brain. However, a precise delineation of the effectively stimulated neuronal populations is notoriously impeded by the widespread and complex distribution of the induced electric field. Here, we propose a method that allows rapid and feasible cortical localization at the individual subject level. The functional relationship between electric field and behavioral effect is quantified by combining experimental data with numerically modeled fields to identify the cortical origin of the modulated effect. Motor evoked potentials (MEPs) from three finger muscles were recorded for a set of random stimulations around the primary motor area. All induced electric fields were nonlinearly regressed against the elicited MEPs to identify their cortical origin. We could distinguish cortical muscle representation with high spatial resolution and localized them primarily on the crowns and rims of the precentral gyrus. A post-hoc analysis revealed exponential convergence of the method with the number of stimulations, yielding a minimum of about 180 random stimulations to obtain stable results. Establishing a functional link between the modulated effect and the underlying mode of action, the induced electric field, is a fundamental step to fully exploit the potential of TMS. In contrast to previous approaches, the presented protocol is particularly easy to implement, fast to apply, and very robust due to the random coil positioning and therefore is suitable for practical and clinical applications.



https://doi.org/10.1016/j.neuroimage.2021.118654
Prokhorova, Alexandra; Ley, Sebastian; Ruiz, Alvaro Yago; Scapaticci, Rosa; Crocco, Lorenzo; Helbig, Marko
Preliminary investigations of microwave imaging algorithms for tissue temperature estimation during hyperthermia treatment. - In: 2021 International Conference on Electromagnetics in Advanced Applications (ICEAA), (2021), S. 079-084

Monitoring of the temperature distribution inside the tissue to be treated plays the key role in success and safety of thermal therapies. Active microwave imaging represents a promising approach for non-invasive tissue temperature monitoring during hyperthermia treatment. In the present paper, an approach for quantitative non-invasive tissue temperature estimation via UWB imaging is described. Two types of imaging algorithms are considered - Delay and Sum beamforming and Truncated Singular Value Decomposition scheme. The capabilities of the proposed imaging algorithms are demonstrated by experiments with liquid phantoms. The results of our investigation show that both imaging algorithms can be applied for detection and estimation of the temperature induced dielectric property changes.



https://doi.org/10.1109/ICEAA52647.2021.9539731
Stricker, Ronny; Aganian, Dustin; Sesselmann, Maximilian; Seichter, Daniel; Engelhardt, Marius; Spielhofer, Roland; Hahn, Matthias; Hautz, Astrid; Debes, Klaus; Groß, Horst-Michael
Road surface segmentation - pixel-perfect distress and object detection for road assessment. - In: 2021 IEEE 17th International Conference on Automation Science and Engineering (CASE), (2021), S. 1789-1796

Visual road assessment, which is carried out by many countries, involves the evaluation of millions of surface images. This exhaustive task is usually done manually and therefore is costly in terms of time and prone to failure. Different methods for automatic distress detection have been presented in the literature recently. However, most of the approaches are focused on crack detection only. This paper focuses on detecting multiple distress types and object classes on asphalt roads, aiming to fully automate distress detection on road surfaces in Austria, Switzerland, and Germany using image segmentation with neural networks. The paper introduces a distress and object catalog developed by experts of the involved countries that guarantees convertibility into federal distress catalogs. We evaluate the performance gain of different neural network architectures and advanced training techniques by conducting extensive experiments.



https://doi.org/10.1109/CASE49439.2021.9551591
Irmak, Hasan; Ziener, Daniel; Alachiotis, Nikolaos
Increasing flexibility of FPGA-based CNN accelerators with dynamic partial reconfiguration. - In: 2021 31st International Conference on Field-Programmable Logic and Applications, (2021), S. 306-311

Convolutional Neural Networks (CNN) are widely used for image classification and have achieved significantly accurate performance in the last decade. However, they require computationally intensive operations for embedded applications. In recent years, FPGA-based CNN accelerators have been proposed to improve energy efficiency and throughput. While dynamic partial reconfiguration (DPR) is increasingly used in CNN accelerators, the performance of dynamically reconfigurable accelerators is usually lower than the performance of pure static FPGA designs. This work presents a dynamically reconfigurable CNN accelerator architecture that does not sacrifice throughput performance or classification accuracy. The proposed accelerator is composed of reconfigurable macroblocks and dynamically utilizes the device resources according to model parameters. Moreover, we devise a novel approach, to the best of our knowledge, to hide the computations of the pooling layers inside the convolutional layers, thereby further improving throughput. Using the proposed architecture and DPR, different CNN architectures can be realized on the same FPGA with optimized throughput and accuracy. The proposed architecture is evaluated by implementing two different LeNet CNN models trained by different datasets and classifying different classes. Experimental results show that the implemented design achieves higher throughput than current LeNet FPGA accelerators.



https://doi.org/10.1109/FPL53798.2021.00061
Seichter, Daniel; Köhler, Mona; Lewandowski, Benjamin; Wengefeld, Tim; Groß, Horst-Michael
Efficient RGB-D semantic segmentation for indoor scene analysis. - In: 2021 IEEE International Conference on Robotics and Automation, (2021), S. 13525-13531

Analyzing scenes thoroughly is crucial for mobile robots acting in different environments. Semantic segmentation can enhance various subsequent tasks, such as (semantically assisted) person perception, (semantic) free space detection, (semantic) mapping, and (semantic) navigation. In this paper, we propose an efficient and robust RGB-D segmentation approach that can be optimized to a high degree using NVIDIA TensorRT and, thus, is well suited as a common initial processing step in a complex system for scene analysis on mobile robots. We show that RGB-D segmentation is superior to processing RGB images solely and that it can still be performed in real time if the network architecture is carefully designed. We evaluate our proposed Efficient Scene Analysis Network (ESANet) on the common indoor datasets NYUv2 and SUNRGB-D and show that we reach state-of-the-art performance while enabling faster inference. Furthermore, our evaluation on the outdoor dataset Cityscapes shows that our approach is suitable for other areas of application as well. Finally, instead of presenting benchmark results only, we also show qualitative results in one of our indoor application scenarios.



https://doi.org/10.1109/ICRA48506.2021.9561675
Madrin, Febby Purnama; Klemm, Matthias; Supriyanto, Eko
Reliability improvement of UWB tracker for hospital asset management system : case study for TEE probe monitoring. - In: The role of AI in health and social revolution in turbulence era, (2021), S. 69-74

With a limited number of workers or staff in the hospital, it is not possible to manually monitor all of the medical device in the hospital. Many medical devices were lost by mistake, many assets went unused because they were not well-stocked, and many assets were destroyed without recognizing it. This will undoubtedly be very negative to hospitals in terms of resources, which are often expensive, and will, of course, diminish the effectiveness and efficiency of medical services. The necessity for hospitals to modernize their technology is apparent. The rapid advancement of technology allows us to overcome these issues, in fact, IoT-based technology is now so advanced that paper-based technology must be gradually phased out. The technology is a real-time location system (RTLS), there are many different ways to implement this technology, one of them is to use Ultra-Wide band (UWB) technology, with this solution, hospitals can track the location of their medical device, as well as other information - this including the Transesophageal Echo (TEE) Probe. DWM1001 is one of the UWB modules that researchers can develop, but its deployment in hospitals still need more research and reliability. This study will address techniques for improving the reliability of anchor mapping and hybrid Wi-Fi solutions as backup solutions.



https://doi.org/10.1109/ICOIACT53268.2021.9563986
Müller, Steffen; Stephan, Benedict; Groß, Horst-Michael
MDP-based motion planning for grasping in dynamic scenarios. - In: 2021 European Conference on Mobile Robots (ECMR), (2021), insges. 8 S.

Path planning for robotic manipulation is a well understood topic as long as the execution of the plan takes place in a static scene. Unfortunately, for applications involving human interaction partners a dynamic obstacle configuration has to be considered. Furthermore, if it comes to grasping objects from a human hand, there is not a single goal position and the optimal grasping configuration may change during the execution of the grasp movement. This makes a continuous re-planning in a loop necessary. Besides efficiency and security concerns, such periodic planning raises the additional requirement of consistency, which is hard to achieve with traditional sampling based planners. We present an online capable planner for continuous control of a robotic grasp task. The planner additionally is able to resolve multiple possible grasp poses and additional goal functions by applying an MDP-like optimization of future rewards. Furthermore, we present a heuristic for setting edges in a probabilistic roadmap graph that improves the connectivity and keeps edge count low.



https://doi.org/10.1109/ECMR50962.2021.9568813
Mohapatra, Sambit; Yogamani, Senthil; Gotzig, Heinrich; Milz, Stefan; Mäder, Patrick
BEVDetNet: Bird's Eye View LiDAR point cloud based real-time 3D object detection for autonomous driving. - In: 2021 IEEE International Intelligent Transportation Systems Conference (ITSC), (2021), S. 2809-2915

3D object detection based on LiDAR point clouds is a crucial module in autonomous driving particularly for long range sensing. Most of the research is focused on achieving higher accuracy and these models are not optimized for deployment on embedded systems from the perspective of latency and power efficiency. For high speed driving scenarios, latency is a crucial parameter as it provides more time to react to dangerous situations. Typically a voxel or point-cloud based 3D convolution approach is utilized for this module. Firstly, they are inefficient on embedded platforms as they are not suitable for efficient parallelization. Secondly, they have a variable runtime due to level of sparsity of the scene which is against the determinism needed in a safety system. In this work, we aim to develop a very low latency algorithm with fixed runtime. We propose a novel semantic segmentation architecture as a single unified model for object center detection using key points, box predictions and orientation prediction using binned classification in a simpler Bird's Eye View (BEV) 2D representation. The proposed architecture can be trivially extended to include semantic segmentation classes like road without any additional computation. The proposed model has a latency of 4 ms on the embedded Nvidia Xavier platform. The model is 5X faster than other top accuracy models with a minimal accuracy degradation of 2% in Average Precision at IoU = 0.5 on KITTI dataset.



https://doi.org/10.1109/ITSC48978.2021.9564490
Sabel, Bernhard A.; Kresinsky, Anton; Cárdenas-Morales, Lizbeth Karina; Haueisen, Jens; Hunold, Alexander; Dannhauer, Moritz; Antal, Andrea
Evaluating current density modeling of non-invasive eye and brain electrical stimulation using phosphene thresholds. - In: IEEE transactions on neural systems and rehabilitation engineering, ISSN 1558-0210, Bd. 29 (2021), S. 2133-2141

https://doi.org/10.1109/TNSRE.2021.3120148