Logo des Repositoriums

Auflistung nach Autor:in "Bracke, Justus"

1 - 2 von 2
  • Konferenzbeitrag
    Comparison of UAV- and mowing machine-mounted LiDAR for grassland canopy height estimation
    (44. GIL - Jahrestagung, Biodiversität fördern durch digitale Landwirtschaft, 2024) Bracke, Justus; Storch, Marcel; Bald, Janis; Jarmer, Thomas
    Towards autonomous process monitoring, canopy height estimation in grassland based on data from a mowing machine-mounted LiDAR and a UAV-LiDAR system is compared to manually measured ground truth heights. In a field trial, a LiDAR mounted on the cabin roof of the mowing machine recorded data during the mowing process, while two recording flights before and after the mowing were conducted with a UAV-LiDAR. The data from both systems were processed similarly and parameters such as height estimation method, spatial resolution and percentile filters were systematically varied to investigate their influence on height estimation accuracy. Statistical evaluation showed that canopy height estimates based on the UAV-LiDAR (R² = 0.89, RMSE = 0.05 m) were more accurate and precise than those based on the mowing machine-mounted LiDAR (R² = 0.51, RMSE = 0.08 m). The influence of the different investigated parameters varied.
  • Konferenzbeitrag
    Evaluating synthetic vs. real data generation for AI-based selective weeding
    (43. GIL-Jahrestagung, Resiliente Agri-Food-Systeme, 2023) Iqbal, Naeem; Bracke, Justus; Elmiger, Anton; Hameed, Hunaid; von Szadkowski, Kai
    Synthetic data has the potential to reduce the cost for ML training in agriculture but poses its own set of problems compared to real data acquisition. In this work, we present two methods of training data acquisition for the application of machine vision algorithms in the use case of selective weeding. Results from ML experiments suggest that current methods for generating synthetic data in the field of agriculture cannot fully replace real data but may greatly reduce the quantity of real data required for model training.