Logo des Repositoriums

Auflistung nach Schlagwort "Biodiversity"

1 - 5 von 5
  • Textdokument
    Automatic Plant Cover Estimation with Convolutional Neural Networks
    (INFORMATIK 2021, 2021) Körschens, Matthias; Bodesheim, Paul; Römermann, Christine; Bucher, Solveig Franziska; Migliavacca, Mirco; Ulrich, Josephine; Denzler, Joachim
    Monitoring the responses of plants to environmental changes is essential for plant biodiversity research. This, however, is currently still being done manually by botanists in the field. This work is very laborious, and the data obtained is, though following a standardized method to estimate plant coverage, usually subjective and has a coarse temporal resolution. To remedy these caveats, we investigate approaches using convolutional neural networks (CNNs) to automatically extract the relevant data from images, focusing on plant community composition and species coverages of 9 herbaceous plant species. To this end, we investigate several standard CNN architectures and different pretraining methods. We find that we outperform our previous approach at higher image resolutions using a custom CNN with a mean absolute error of 5.16%. In addition to these investigations, we also conduct an error analysis based on the temporal aspect of the plant cover images. This analysis gives insight into where problems for automatic approaches lie, like occlusion and likely misclassifications caused by temporal changes.
  • Konferenzbeitrag
    Interactive Data Exploration for Geoscience
    (Datenbanksysteme für Business, Technologie und Web (BTW 2017) - Workshopband, 2017) Beilschmidt, Christian; Drönner, Johannes; Mattig, Michael; Schmidt, Marco; Authmann, Christian; Niamir, Aidin; Hickler, Thomas; Seeger, Bernhard
    Data-driven research requires interactive systems supporting fast and intuitive data ex- ploration. An important component is the user interface that facilitates this process. In biodiversity research, data is commonly of spatio-temporal nature. This poses unique opportunities for visual analytics approaches. In this paper we present the core concepts of the web-based front end of our VAT (Visualization, Analysis and Transformation) system, a distributed geo-processing application. We present the results of a user study and highlight unique features for the management of time and the generalization of data.
  • Konferenzbeitrag
    Semantic Search for Biological Datasets: A Usability Study on Modes of Querying and Explaining Search Results
    (BTW 2023, 2023) Löffler, Felicitas; Shafiei, Fateme; Witte, René; König-Ries, Birgitta; Klan, Friederike
    Dataset discovery is a frequent task in daily research practice, yet studies are missing that explore the usability of user interfaces (UI) in data portals. In particular, very few user studies exist that analyze whether particular elements in the user interface are useful for search tasks. We aim to address those needs for more specific usability evaluations in dataset search. In this work, wepresent a flexible semantic search over biological datasets with two user interfaces. The search result contains semantically related terms, such as synonyms or more specific terms, obtained from domain ontologies. We evaluated the system in a user study with 20 scholars. We focused on two components, the query input to explore a search in categories (entity types) in comparision to a single input field, and we analyzed textual highlightings in the returned datasets to study whether users are distracted by semantic information such as URIs. Our results show that users prefer interfaces with a single input field for search tasks they are not familiar with, and that users appreciate explanations with terminologies and URIs.
  • Textdokument
    Towards a Semantic Toolbox for Reproducible Knowledge Graph Generation in the Biodiversity Domain - How to Make the Most out of Biodiversity Data
    (INFORMATIK 2021, 2021) Babalou, Samira; Schellenberger Costa, David; Kattge, Jens; Römermann, Christine; König-Ries, Birgitta
    Knowledge Graphs are widely regarded as one of the most promising ways to manage and link information in the age of Big Data. Their broad uptake is still hindered by the large effort required to create and maintain them, though. In this paper, we propose the creation of a semantic toolbox that will support data owners in transforming their databases into reproducible, dynamically extendable knowledge graphs that can be integrated and jointly used. We showcase the need, potential benefits and first steps towards the solution in our example domain, biodiversity research.
  • Zeitschriftenartikel
    VAT: A Scientific Toolbox for Interactive Geodata Exploration
    (Datenbank-Spektrum: Vol. 17, No. 3, 2017) Beilschmidt, Christian; Drönner, Johannes; Mattig, Michael; Schmidt, Marco; Authmann, Christian; Niamir, Aidin; Hickler, Thomas; Seeger, Bernhard
    Data-driven research requires interactive systems supporting fast and intuitive data exploration. An important component is the user interface that facilitates this process. In biodiversity research, data is commonly of spatio-temporal nature. This poses unique opportunities for visual analytics approaches. In this paper we present the core concepts of the web-based front end of our vat (Visualization, Analysis and Transformation) system, a distributed geo-processing application. We present the results of two user studies and highlight unique features, among others for the management of time and the generalization of data.