Fraunhofer-Institut für Optronik, Systemtechnik und Bildauswertung IOSB

Filters

2 1
Reset filters

Settings
Now showing 1 - 3 of 3
  • Publication
    Sensor-based characterization of construction and demolition waste at high occupancy densities using synthetic training data and deep learning
    ( 2024)
    Kronenwett, Felix
    ;
    Maier, Georg
    ;
    Leiss, Norbert
    ;
    Gruna, Robin
    ;
    Thome, Volker
    ;
    Längle, Thomas
    Sensor-based monitoring of construction and demolition waste (CDW) streams plays an important role in recycling (RC). Extracted knowledge about the composition of a material stream helps identifying RC paths, optimizing processing plants and form the basis for sorting. To enable economical use, it is necessary to ensure robust detection of individual objects even with high material throughput. Conventional algorithms struggle with resulting high occupancy densities and object overlap, making deep learning object detection methods more promising. In this study, different deep learning architectures for object detection (Region-based CNN/Region-based Convolutional Neural Network (Faster R-CNN), You only look once (YOLOv3), Single Shot MultiBox Detector (SSD)) are investigated with respect to their suitability for CDW characterization. A mixture of brick and sand-lime brick is considered as an exemplary waste stream. Particular attention is paid to detection performance with increasing occupancy density and particle overlap. A method for the generation of synthetic training images is presented, which avoids time-consuming manual labelling. By testing the models trained on synthetic data on real images, the success of the method is demonstrated. Requirements for synthetic training data composition, potential improvements and simplifications of different architecture approaches are discussed based on the characteristic of the detection task. In addition, the required inference time of the presented models is investigated to ensure their suitability for use under real-time conditions.
  • Publication
    Making use of climate information for sustainable preservation of cultural heritage: applications to the KERES project
    ( 2023)
    Kotova, Lola
    ;
    Leissner, Johanna
    ;
    Winkler, Matthias
    ;
    Kilian, Ralf
    ;
    Bichlmair, Stefan
    ;
    Antretter, Florian
    ;
    Moßgraber, Jürgen
    ;
    Reuter, Jürgen orcid-logo
    ;
    Hellmund, Tobias
    ;
    Matheja, Katharina
    ;
    Rohde, Michael
    ;
    Mikolajewicz, Uwe
    According to the final report of the European Union OMC expert group on strengthening cultural heritage resilience for anthropogenic climate change, the impacts of climate change, particularly extreme weather events, on cultural heritage in Europe have become increasingly evident in recent years and are progressing at an unprecedented speed and scale. Archaeological sites, museum collections, and historical buildings and structures are affected, among others, by rising temperatures or by heavy storms and precipitation events. Deep scientific knowledge about future climate projections is required to develop appropriate preservation strategies and measures to protect and adapt cultural heritage. In this paper we present the first set of results of the KERES project. The project focuses on the impacts of future extreme climate events on the built heritage and historic gardens. An ensemble of climate simulations is used to analyze changes in both climatology and extreme events for several climate variables at two cultural heritage sites in Germany. In this study, a methodology was developed to guide climate scientists on how to better tailor climate information for the needs of stakeholders in the cultural heritage sector. It would help the stakeholders to integrate the results of climate projections into the prevention and emergency management, in particular for the risk assessment of extreme events. The effects of interpolation from a model grid to a location of cultural heritage site and advantages of an ensemble approach have been demonstrated in the study.
  • Publication
    BauCycle - Verwertungsstrategie für feinkörnigen Bauschutt
    ( 2018)
    Dittrich, Sebastian
    ;
    Thome, Volker
    ;
    Nühlen, Jochen
    ;
    Gruna, Robin
    ;
    Dörmann, Joseph
    Das Fraunhofer-interne Projekt "BauCycle" entwickelt eine ganzheitliche Verwertungsstrategie für feinkörnigen Bauschutt, welcher bisher meist ungenutzt auf Deponien verbracht wird. In einem ersten Schritt wurde eine Methodik zur "chemischen" Sortierung von Bauschutt kleiner 2 mm entwickelt um Störstoffe wie Gips selektiv aus Beton oder Kalksandstein austragen zu können. Basierend auf den Sortiermöglichkeiten fand die Entwicklung von Bauprodukten mit einem signifikanten Anteil an Bauschutt statt. Es konnte ein Porenbetonstein hergestellt werden bei dem der Primärrohstoff Sand zu 30 Ma.-Prozent durch feinkörniges Bauschuttmaterial ersetzt werden konnte. Zudem wurde ein Akustikputz für Innenanwendungen entwickelt, welcher absolut vergleichbar zu marktüblichen Produkten ist. Neben der Entwicklung neuer Produkte steht auch die Betrachtung geltender Richtlinien im Hinblick auf den Einsatz von neuentwickelten Baustoffen sowie deren ökologische Wirkung im Fokus. Dazu wurden relevante Indikatoren wie Rohstoffqualität und -verfügbarkeit definiert und für eine Modellregion dargestellt. Dies geschah nicht zuletzt unter Berücksichtigung logistischer Parameter. Der finale Schritt im Projekt ist die Implementierung einer Handelsplattform über welche Bauschuttfraktionen bedarfsgerecht angeboten bzw. nachgefragt werden können.