Logo des Repositoriums

Auflistung nach Schlagwort "digital agriculture"

1 - 3 von 3
  • Konferenzbeitrag
    Data sovereignty needs in agricultural use cases
    (43. GIL-Jahrestagung, Resiliente Agri-Food-Systeme, 2023) Boye, Frederik; Matar, Raghad; Neuschwander, Philipp
    With today's shift towards digital farming, data transfer and processing are becoming increasingly essential for optimizing farm operations. In the complex arable farming sector, there exist numerous use cases involving a variety of actors and processes. Designing new digital ecosystems, such as agricultural data spaces that support existing workflows while enabling new opportunities for data-driven services, requires a good understanding of existing processes and data sovereignty needs. In this paper, we categorize data exchange use cases in arable farming and analyze the respective data sovereignty needs for each category. The results of our contribution can be used as a basis for further analysis and evaluation of data-sharing approaches in terms of their suitability for meeting different data sovereignty needs in agriculture, as well as in the process of requirements analysis when designing such systems.
  • Konferenzbeitrag
    Development and field evaluation of a multichannel LoRa sensor for IoT monitoring in berry orchards
    (41. GIL-Jahrestagung, Informations- und Kommunikationstechnologie in kritischen Zeiten, 2021) Shamshiri, Redmond R.; Weltzien, Cornelia
    Evaluation of long-range wireless transceivers with respect to their power consumption, network connectivity, and coverage under extreme field conditions is necessary prior to their deployment in large-scale commercial orchards. This paper reports on the development and field performance of an affordable multi-channel wireless data acquisition for IoT monitoring of environmental variations in berry orchards. A connectivity board was custom-designed based on the powerful dual-core 32-bit microcontroller with WiFi antenna and LoRa modulation at 868 MHz. The objective was to verify the possibility of transmitting multiple sensor readings with lower power consumption while increasing the reliability and stability of wireless communication at long distances (over 1.7 km). Collected data from the wireless sensor was compared and found to be consistent with measurements of a data logger installed in the same locations. The presented paper highlights the advantages of LoRa sensors for digital agriculture and the experience in real-time monitoring of environmental parameters in berry orchards.
  • Konferenzbeitrag
    Indicator plant species detection in grassland using EfficientDet object detector
    (42. GIL-Jahrestagung, Künstliche Intelligenz in der Agrar- und Ernährungswirtschaft, 2022) Basavegowda, Deepak Hanike; Mosebach, Paul; Schleip, Inga; Weltzien, Cornelia
    Extensively used grasslands (meadows and pastures) are ecologically valuable areas in the agricultural landscape and part of the multifunctional agriculture. In Germany, the quality of these grasslands is assessed based on the occurrence of certain plant species known as indicator or character species, with indicators being defined at regional level. Therefore, the recognition of these indicators on a spatial level is a prerequisite for monitoring grassland biodiversity. The identification of indicator species for the status quo of grassland using traditional methods was found to be challenging and tedious. Deep learning-algorithms applied to high-resolution UAV imagery could be the key solution, where UAV with remote sensors can map a large area of grassland in comparison to manual or ground mapping methods and deep learning-algorithms can automate the detection process. In this research work, we use an EfficientDet based algorithm to train an object detection model capable of recognizing indicators on RGB data. The experimental results show that this approach is very promising in contrast to the difficult and time-consuming manual recognition methods. The model was trained with the momentum-SGD optimizer with a momentum value of 0.9 and a learning rate of 0.0001. The model was trained and tested on 1200 images and achieves 45.7 AP (and 85.7 AP50) on test data set. The dataset includes images of four distinct indicator plant species: Armeria maritima, Campanula patula, Cirsium oleraceum, and Daucus carota