A5020996505- Remote Sensing and LiDAR Applications
- Satellite Image Processing and Photogrammetry
- Robotics and Sensor-Based Localization
- Automated Road and Building Extraction
- Video Surveillance and Tracking Methods
- Remote Sensing in Agriculture
- 3D Surveying and Cultural Heritage
- Advanced Neural Network Applications
- Remote-Sensing Image Classification
- Advanced Image and Video Retrieval Techniques
- Autonomous Vehicle Technology and Safety
- Infrared Target Detection Methodologies
- Image and Object Detection Techniques
- Geochemistry and Geologic Mapping
- Advanced Measurement and Detection Methods
- Advanced Optical Sensing Technologies
- Smart Parking Systems Research
- Advanced Vision and Imaging
- Geographic Information Systems Studies
- Advanced Image Fusion Techniques
- Impact of Light on Environment and Health
- Land Use and Ecosystem Services
- Inertial Sensor and Navigation
- Spectroscopy and Chemometric Analyses
- Advanced Computational Techniques and Applications
Deutsches Zentrum für Luft- und Raumfahrt e. V. (DLR)
2015-2024
Remote Sensing Solutions (Germany)
2012
Zimmer Biomet (Germany)
2002
Institute for Environment and Human Security
2001
Technical University of Munich
2000
Vehicle detection has been an important research field for years as there are a lot of valuable applications, ranging from support traffic planners to real-time management. Especially cars in dense urban areas is interest due the high volume and limited space. In city many car-like objects (e.g., dormers) appear which might lead confusion. Additionally, inaccuracy road databases supporting extraction process be handled proper way. This paper describes integrated processing chain utilizes...
Given that ground stationary infrastructures for traffic monitoring are barely able to handle everyday volumes, there is a risk they could fail altogether in situations arising from mass events or disasters. In this work, we present an alternative approach during disaster and events, which based on airborne optical sensor system. With system, image sequences automatically examined board aircraft estimate road information, such as vehicle positions, velocities driving directions. The...
Abstract. In Earth observation, multimodal data fusion is an intuitive strategy to break the limitation of individual data. Complementary physical contents sources allow comprehensive and precise information retrieval. With current satellite missions, such as ESA Copernicus programme, various will be accessible at affordable cost. Future applications have many options for sources. Such a privilege can beneficial only if algorithms are ready work with However, studies mostly focus on two...
Advances in remote sensing image processing techniques have further increased the demand for annotated datasets. However, preparing multi-temporal 2D/3D multimodal data is especially challenging, both costs of annotation step and lack acquisitions available on same area. We introduce Simulated Multimodal Aerial Remote Sensing (SMARS) dataset, a synthetic dataset aimed at tasks urban semantic segmentation, change detection, building extraction, along with description pipeline to generate them...
Recent years have witnessed the fast development of UAVs (unmanned aerial vehicles). As an alternative to traditional image acquisition methods, bridge gap between terrestrial and airborne photogrammetry enable flexible high resolution images. However, georeferencing accuracy is still limited by low-performance on-board GNSS INS. This paper investigates automatic geo-registration individual UAV or blocks matching image(s) with a previously taken georeferenced image, such as satellite height...
Multi-class vehicle detection from airborne imagery with orientation estimation is an important task in the near and remote vision domains applications traffic monitoring disaster management. In last decade, we have witnessed significant progress object ground imagery, but it still its infancy mostly due to scarcity of diverse large-scale datasets. Despite being a useful tool for different applications, current datasets only partially reflect challenges real-world scenarios. To address this...
The environmental perception of an autonomous vehicle is limited by its physical sensor ranges and algorithmic performance, as well occlusions that degrade understanding ongoing traffic situation. This not only poses a significant threat to safety limits driving speeds, but it can also lead inconvenient maneuvers. Intelligent Infrastructure Systems help alleviate these problems. An System fill in the gaps vehicle’s extend field view providing additional detailed information about...
Flow, a state of deep task engagement, is associated with optimal experience and well-being, making its detection prolific HCI research focus. While physiological sensors show promise for flow detection, most studies are lab-based. Furthermore, brain sensing during natural work remains unexplored due to the intrusive nature traditional EEG setups. This study addresses this gap by using wearable, around-the-ear observe knowledge work, measuring throughout an entire day. In semi-controlled...
Building footprint information is vital for 3D building modeling. Traditionally, in remote sensing, footprints are extracted and delineated from aerial imagery and/or LiDAR point cloud. Taking a different approach, this paper dedicated to the optimization of OpenStreetMap (OSM) exploiting contour information, which derived deep learning-based semantic segmentation oblique images acquired by Unmanned Aerial Vehicle (UAV). First, simplified model Level Detail 1 (LoD 1) initialized using OSM...
Abstract Purpose Large area traffic monitoring with high spatial and temporal resolution is a challenge that cannot be served by today available static infrastructure. Therefore, we present an automatic near real-time approach using data of airborne digital camera system frame rate up to 3 fps. Methods By performing direct georeferencing on the obtained aerial images use GPS/IMU are able conduct extraction. The processor consists mainly three steps which road extraction supported priori...
Die Beobachtung von Naturkatastrophen, Grosereignissen und Unfallen mit flugzeuggestutzten optischen Sensoren in Echtzeit ist ein derzeit wichtiges Thema Forschung Entwicklung. In diesem Zusammenhang wird die Leistungsfahigkeit preisgunstigen Kamerasystemen fur Echtzeitanwendungen Hinblick auf geometrische Genauigkeit, radiometrische Eigenschaften Prozessierungszeiten evaluiert. Der Schwerpunkt liegt bei der Analyse geometrischen Stabilitat Kameras im langjahrigen Betrieb den Grenzen...
The digital transformation taking place in all areas of life has led to a massive increase data – particular, related the places where and ways how we live. To facilitate an exploration new opportunities arising from this development Urban Thematic Exploitation Platform (U-TEP) been set-up. This enabling instrument represents virtual environment that combines open access multi-source repositories with dedicated processing, analysis visualisation functionalities. Moreover, it includes...
A new approach for the traffic congestion detection in time series of optical digital camera images is proposed. It well suited to derive various parameters such as vehicle density, average velocity, beginning and end congestion, length or other monitoring applications. The method based on road segment by change between two with a short lag, usage priori information data base, sizes simple linear model spacing vehicles. estimated velocity profiles experimental acquired airborne remote...
The tremendous advances in deep neural networks have demonstrated the superiority of learning techniques for applications such as object recognition or image classification. Nevertheless, learning-based methods usually require a large amount training data, which mainly comes from manual annotation and is quite labor-intensive. In order to reduce work required generating enough we hereby propose leverage existing labeled data generate annotations automatically. Specifically, pixel labels are...
The 3D information of road infrastructures is growing in importance with the development autonomous driving. In this context, exact 2D position markings as well height play an important role in, e.g., lane-accurate self-localization vehicles. paper, overall task divided into automatic segmentation followed by a refined reconstruction. For task, we applied wavelet-enhanced fully convolutional network on multiview high-resolution aerial imagery. Based resulting segments original images,...
Individual tree crown (ITC) segmentation supports numerous applications in forest management and ecology. In the latter context, special attention is dedicated to study of angular reflection effects, caused by interaction incident sunlight with a canopy. High precision airborne analysis these effects requires multi-view sensor systems ITC segmentation. particular oblique view image difficult has been addressed template based methods. This contribution identifies persistent shortcomings state...
Abstract. Emerging traffic management technologies, smart parking applications, together with transport researchers and urban planners are interested in fine-grained data on space cities. However, there no standardized, complete up-to-date databases for many areas. Moreover, manual collection is expensive time-consuming. Aerial imagery of entire cities can be used to inventory not only publicly accessible dedicated lots, but also roadside areas those private property. For a realistic...
A new model based approach for the traffic congestion detection in time series of airborne optical digital camera images is proposed. It on estimation average vehicle speed road segments. The method puts various techniques together: segments by change between two with a short lag, usage priori information such as data base, sizes and parameters simple linear spacing vehicles. estimated profiles from experimental acquired an sensor - 3K system coincide well reference measurements....
This article describes the workflow of classification algorithm which ranked at 2 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">nd</sup> place in 2018 GRSS Data Fusion Contest. The objective contest was to provide a map with 20 classes on complex urban scenario. available multi-modal data were acquired from hyperspectral, LiDAR and very high-resolution RGB sensors flown same platform over city Houston, TX, USA. obtained by merging deep...
We present a new dataset for development, benchmarking, and evaluation of remote sensing earth observation approaches with special focus on converging perspectives. In order to provide data different modalities, we observed the same scene using satellites, airplanes, unmanned aerial vehicles (UAV), smartphones. The is further complemented by ground-truth information baseline results application scenarios. provided can be freely used anybody interested in will continuously augmented updated.