A5016532646- Embedded Systems Design Techniques
- Parallel Computing and Optimization Techniques
- Interconnection Networks and Systems
- Advanced Vision and Imaging
- Advanced SAR Imaging Techniques
- Low-power high-performance VLSI design
- Speech and Audio Processing
- Robotics and Sensor-Based Localization
- Synthetic Aperture Radar (SAR) Applications and Techniques
- CCD and CMOS Imaging Sensors
- VLSI and Analog Circuit Testing
- Advanced Image Processing Techniques
- Radiation Effects in Electronics
- Advanced Image and Video Retrieval Techniques
- Advanced Adaptive Filtering Techniques
- Video Coding and Compression Technologies
- Advanced Neural Network Applications
- Electrospun Nanofibers in Biomedical Applications
- Radar Systems and Signal Processing
- Numerical Methods and Algorithms
- 3D Printing in Biomedical Research
- Inertial Sensor and Navigation
- Music and Audio Processing
- Image Processing Techniques and Applications
- Tissue Engineering and Regenerative Medicine
Leibniz University Hannover
2016-2025
Hearing4all
2015-2024
Institut für Mikroelektronik- und Mechatronik-Systeme
2010-2021
University of Duisburg-Essen
2013
Karlsruhe Institute of Technology
2013
Leibniz University of Applied Sciences
2012
TU Dortmund University
1996-2011
RWTH Aachen University
1999-2009
Ofcom
2006
Westfälische Hochschule
2003
This paper describes a new architecture and the corresponding implementation of stereo vision system that covers entire process including noise reduction, rectification, disparity estimation, visualization. Dense estimation is performed using non-parametric rank transform semi-global matching (SGM), which among top performing methods outperforms locally-based in terms quality maps robustness under difficult imaging conditions. Stream-based processing SGM despite its non-scan-aligned, complex...
Microgravity eases several constraints limiting experiments with ultracold and condensed atoms on ground. It enables extended times of flight without suspension eliminates the gravitational sag for trapped atoms. These advantages motivated numerous initiatives to adapt operate experimental setups microgravity platforms. We describe design payload, motivations choices, capabilities Bose-Einstein Condensate Cold Atom Laboratory (BECCAL), a NASA-DLR collaboration. BECCAL builds heritage...
Widespread vaccination in pursuit of herd immunity has been recognized as the most promising approach to ending global pandemic coronavirus disease 19 (COVID-19). The children and adolescents extensively debated first COVID-19 vaccine is now approved European countries for aged > 12 years age. Our study investigates hesitancy a cohort German secondary school students. We assessed 903 students between age 9 20 period 17 May 2021 30 June 2021. 68.3% (n = 617) reported intention undergo...
This paper presents novel hybrid architectures that combine grid- and point-based processing to improve the detection performance orientation estimation of radar-based object networks. Purely grid-based models operate on a bird's-eye-view (BEV) projection input point cloud. These approaches suffer from loss detailed information through discrete grid resolution. applies in particular radar detection, where relatively coarse resolutions are commonly used account for sparsity clouds. In...
A Residue Number System (RNS) represents integers through a set of residues obtained by integer division using predefined pairwise coprime moduli. RNS is well-known for enabling efficient carry-free arithmetic and representing large numbers with multiple shorter numerical values. By incorporating additional moduli, the system can be extended into an over-determined Redundant (RRNS), introducing redundancy that facilitates detection correction bit errors.
This paper presents the design, implementation and evaluation of new parallelization schemes for performing dense disparity estimation based on non-parametric rank transform semi-global matching Graphics Processing Units (GPUs). A detailed analysis performance limitating factors (memory throughput, instruction etc.) each part parallel is performed. Thus, a highly optimized mapping scheme onto resources GPU obtained. The resulting performs at 27 frames per second 1024×768 pixel images with...
The performance of automotive radar systems is expected to significantly increase in the near future. With enhanced resolution capabilities more accurate and denser point clouds traffic participants roadside infrastructure can be acquired so amount gathered information growing drastically. One main driver for this development global trend towards self-driving cars, which all rely on precise fine-grained sensor information. New signal processing concepts have developed order provide...
The background of this paper is to apply advanced real-time gait analysis walking interventions in daily life setting. A vast wearable devices provide information but not more than pedometer functions such as step counting, displacement, and velocity. This suggests a method based on head-worn inertial measurement unit. novel implements detection events (heel strike, toe off, mid-stance phase) immediately provides detailed spatiotemporal parameters. reliability was proven by with over 11 000...
Lipoprotein(a) (Lp(a)) is a genetic risk factor for cardiovascular disease (CVD) and associated with the induction sustaining of atherosclerotic diseases (ASCVD). Since 2008 Lp(a) along progressive CVD has been approved as an indication regular lipoprotein apheresis (LA) in Germany. The German Lipoprotein Apheresis Registry (GLAR) initiated to provide statistical evidence assessment extracorporeal procedures treat dyslipidemia both LDL-cholesterol (LDL-C) Lp(a). GLAR now allows prospective...
Abstract Scholem's powers are primarily exegetical. An original thinker he's not. But the object of his life-long exegesis, namely Kabbalah, and messianic mystical tradition within Judaism, allows for insight. For instance, this from chapter one, "With Gershom Scholem: Interview”: “The kabbalist has a fundamental feeling that there is mysteri—a secret—in world… If humanity should ever lose world, then it's all over with us. I don't believe we'll come to that. position not only mystery in...
Abstract. The stereo matching method semi-global (SGM) relies on consistency constraints during the cost aggregation which are enforced by so-called penalty terms. This paper proposes new and evaluates four functions for SGM. Due to mutual dependencies, two types of calculation, census rank transform, considered. Performance is measured using original degenerated images exhibiting radiometric changes noise from Middlebury benchmark. best performing inversely proportional negatively linear...
Pedestrian detection will play a major role in future driver assistance and autonomous driving. One powerful algorithm this field uses HOG features to describe the specific properties of pedestrians images. To determine their locations, are extracted classified window-wise from different scales an input image. The results classification finally merged remove overlapping detections. real-time execution method requires FPGA- or ASIC-architectures. Recent work focused on accelerating feature...
Despite lockdown measures, intense symptom-based PCR, and antigen testing, the SARS-CoV-2 pandemic spread further. In this open observational study conducted in Lower Saxony, Germany, voluntary PCR tests were performed from April 2020 until June 2021, supported by serum antibody testing to prove whether subjects with none or few symptoms of COVID-19 is a suitable tool manage pandemic. different mobile stations, 4,817 three working fields participated testing. Serum screening using ViraChip...
This paper presents a method to learn the Cartesian velocity of objects using an object detection network on automotive radar data. The proposed is self-supervised in terms generating its own training signal for velocities. Labels are only required single-frame, oriented bounding boxes (OBBs). velocities or contiguous sequences, which expensive obtain, not required. general idea pre-train without single-frame OBB labels, and then exploit network's predictions unlabelled data training. In...
Modern and future AI-based automotive applications, such as autonomous driving, require the efficient real-time processing of huge amounts data from different sensors, like camera, radar, LiDAR. In ZuSE-KI-AVF project, multiple university, industry partners collaborate to develop a novel massive parallel processor architecture, based on cus-tomized RISC-V host processor, an high-performance vertical vector coprocessor. addition, software development framework is also provided efficiently...
Abstract We report on the design and construction of a sounding rocket payload capable performing atom interferometry with Bose-Einstein condensates $$^{41}$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:msup><mml:mrow/><mml:mn>41</mml:mn></mml:msup></mml:math> K $$^{87}$$ xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:msup><mml:mrow/><mml:mn>87</mml:mn></mml:msup></mml:math> Rb. The apparatus is designed to be launched in two consecutive missions VSB-30 qualified...
Processor performance is often limited by cache efficiency, a critical aspect in any signal processing task. A RISC-V processor combined with the presented this paper will be part of deep drilling component manufactured to further automate entire process. Therefore, acquisition, storage, and in-situ are required. This presents data optimization for high designed operate under extreme conditions up 175°C 200MPa while maintaining stable frequency 180MHz. At heart new design XFAB's XT018...
The exploration of the design space for heterogeneous reconfigurable systems on chip (SoC) becomes more and important. As modern SoCs include a variety different architecture blocks, ensuring flexibility as well highest performance, it is mandatory to prune in an early stage process order achieve short innovation cycles new products. Therefore, goal this work provide estimations implementation specific parameters like throughput rate, power dissipation silicon area by means cost functions. A...