Logo des Repositoriums

Auflistung nach Schlagwort "Agriculture"

1 - 4 von 4
  • Konferenzbeitrag
    A Core Ontology to Support Agricultural Data Interoperability
    (BTW 2023, 2023) Abdelmageed, Aly; Hatem, Shahenda; ael, Tasneem; Medhat, Walaa; König-Ries, Birgitta; Ellakwa, Susan; Elkafrawy, Passent; Algergawy, Alsayed
    The amount and variety of raw data generated in the agriculture sector from numeroussources, including soil sensors and local weather stations, are proliferating. However, these raw data in themselves are meaningless and isolated and, therefore, may offer little value to the farmer. Data usefulness is determined by its context and meaning and by how it is interoperable with data from other sources. Semantic web technology can provide context and meaning to data and its aggregation by providing standard data interchange formats and description languages. In this paper, we introduce the design and overall description of a core ontology that facilitates the process of data interoperability in the agricultural domain.
  • Zeitschriftenartikel
    Objekterkennung im Weinanbau – Eine Fallstudie zur Unterstützung von Winzertätigkeiten mithilfe von Deep Learning
    (HMD Praxis der Wirtschaftsinformatik: Vol. 56, No. 5, 2019) Heinrich, Kai; Zschech, Patrick; Möller, Björn; Breithaupt, Lukas; Maresch, Johannes
    Die voranschreitende Digitalisierung revolutioniert sämtliche Wirtschaftszweige und bringt somit auch langfristige Veränderungen für den landwirtschaftlichen Sektor mit sich, wo auf Basis intelligenter Informationssysteme zahlreiche Daten gesammelt und im Zuge neuer Geschäftsmodelle ausgewertet werden. Vor diesem Hintergrund präsentiert der vorliegende Beitrag eine Big-Data-Analytics-Fallstudie aus dem Bereich des Weinanbaus, wo mithilfe von mobilen Aufnahmegeräten umfangreiches Bildmaterial aufgezeichnet wurde, um eine automatisierte Objekterkennung zur Unterstützung von operativen Winzertätigkeiten, wie zum Beispiel das Zählen von Reben, die Identifikation von Rebfehlstellen oder die Prognose von potentiellem Erntegut, realisieren zu können. Hierbei bestand die Herausforderung unter anderem darin, landwirtschaftlich relevante Weinobjekte wie Reben, Trauben und Beeren über die einzelnen Hierarchieebenen hinweg erkennen zu können und diese auch in Bezug auf bewegtes Bildmaterial folgerichtig zu zählen. Zur Bewältigung derartiger Herausforderungen werden einige Lösungsansätze vorgestellt, die auf modernen Deep-Learning-Verfahren der bildbasierten Objekterkennung basieren. Der Beitrag wird abgerundet mit einer Diskussion und Implikationen für analytische Anwendungen in der landwirtschaftlichen Praxis. The transformation towards a digitized world introduces major changes to all economic sectors, among them the sector of agriculture, where intelligent information systems help to gather and analyze vast amounts of data to provide new business functions and models. Given this background, this article describes a big data analytics case study from the field of viticulture, where extensive image material was recorded using mobile recording devices in order to implement automated object detection to support operational vineyard activities, such as counting vines, identifying missing plants or predicting potential harvests. One of the challenges here was to correctly identify relevant wine objects such as vines, grapes and berries across their different hierarchical levels and to consistently count them in relation to moving image material. The authors provide a solution to those challenges by designing a data analysis process based on a deep learning framework for object detection. Additionally, the results as well as implications for the application of the proposed models in the field of agrarian management are discussed at the end of the article.
  • Zeitschriftenartikel
    Ontology-Based Mobile Communication in Agriculture
    (KI - Künstliche Intelligenz: Vol. 27, No. 4, 2013) Grimnes, Gunnar Aastrand; Kiesel, Malte; Bernardi, Ansgar
    This paper describes the use of semantic technologies to enable a public/private communication network in the iGreen project. The motivation for using semantic technologies is outlined, and a description of the iGreen ontology-server is given, and the services this provides to users and developers. We discuss the semantic data-sets published in iGreen and the steps taken to enrich and interlink these.
  • Konferenzbeitrag
    State of the Art Open Access Remote Sensing with ESA Sentinel 1 SAR Data
    (INFORMATIK 2023 - Designing Futures: Zukünfte gestalten, 2023) McClelland, Jennifer; Riedel, Tanja; Beyer, Florian; Gerighausen, Heike; Golla, Burkhard
    Tackling the consequences of climate change has become a global issue. Climate change will clearly influence our common lifestyle enormously in near future. This involves increasingly frequent sudden weather changes and extreme temperatures as well as drastic changes in water quality and availability. Because of our constant growing global population, nutritional habits and agricultural practices, the share of the agricultural impact on global anthropogenic greenhouse gas emissions take on an estimated 10-12 %. At the same time, fulfilling the agricultural demand is becoming increasingly challenging due to unpredictable farming conditions. Without immediate collaborative efforts including focused research, employment and adaption of state of the art technologies, this issue will not be tackled soon enough, to avoid massive limitations and enormous losses. A very promising large-scale technology to monitor agricultural ecosystems and activities is by means of earth observation imagery derived by Synthetic Aperture Radar (SAR). Radar backscatter e.g. allows insights to crop conditions, soil properties and direct mapping of vegetation growth. Open access technologies offer the best solutions for collaborative efforts, thus minimising financial and legal constraints in comparison to technologies residing in the commercial sector. Here, we combine and build on state-of-the-art tools and technologies to provide an easy to employ Sentinel-1 SAR pre-processing tool as well as a Germany wide, open access, pre-processed, analysis- ready database of Sentinel-1 SAR data. All tools used and developed are open source and freely available. With the employment of modern software developing methods and tools for a scalable and maintainable architecture, these products can be easily extended and adapted. By deployment of up to date machine learning methods, combining the resulting datasets with other relevant parameters, not to say the least, e.g. early prediction of optimal sowing, harvesting and fertilisation time points can be determined as well as many more valuable insights for successful, resource-efficient and environmentally friendly farming. Furthermore, the pre-processing of SAR datasets is not only substantial for the field of agriculture but for a wide range of other fields concerning environmental observations.