Logo des Repositoriums

Auflistung nach Schlagwort "data annotation"

1 - 3 von 3
  • Konferenzbeitrag
    Active-learning-driven deep interactive segmentation for cost-effective labeling of crop-weed image data
    (43. GIL-Jahrestagung, Resiliente Agri-Food-Systeme, 2023) Sikouonmeu, Freddy; Atzmueller, Martin
    Active learning has shown its reliability in (semi-)supervised machine learning tasks to reduce the labeling cost for large datasets in various areas. However, in the agricultural field, despite past attempts to reduce the labeling cost and the burden on the labeler in acquiring image labels, the load during the acquisition of pixel-level labels for semantic image segmentation tasks remains high. Typically, the respective pixel-level masks are acquired manually by drawing polygons over irregular and complex-shaped object boundaries. In contrast, this paper proposes a method leveraging the power of a click-based deep interactive segmentation model (DISEG) in an active learning approach to harvest high-quality image segmentation labels at a low cost for training a real-time task model by only clicking on the objects’ fore- and background surfaces. Our first experimental results indicate that with an average of 3 clicks per image object and using only 3% of the unlabeled dataset, we can acquire pixel-level labels with good quality at low cost.
  • Konferenzbeitrag
    AI-supported data annotation in the context of UAV-based weed detection in sugar beet fields using Deep Neural Networks
    (42. GIL-Jahrestagung, Künstliche Intelligenz in der Agrar- und Ernährungswirtschaft, 2022) Jonas Boysen, Jonas; Stein, Anthony
    Recent Deep Learning-based Computer Vision methods proved quite successful in various tasks, also involving the classification, detection and segmentation of crop and weed plants with Convolutional Neural Networks (CNNs). Such solutions require a vast amount of labeled data. The annotation is a tedious and time-consuming task, which often constitutes a limiting factor in the Machine Learning process. In this work, an approach for an annotation pipeline for UAV-based images of sugar beet fields of BBCH-scale 12 to 17 is presented. For the creation of pixel-wise annotated data, we utilize a threshold-based method for the creation of a binary plant mask, a row detection based on Hough Transform and a lightweight CNN for the classification of small, cropped images. Our findings demonstrate that an increased image data annotation efficiency can be reached by using an AI approach already at the crucial Machine Learning-process step of training data collection.
  • Konferenzbeitrag
    Variability of annotations over time: An experimental study in the dementia-related named entity recognition domain
    (INFORMATIK 2024, 2024) Stoev, Teodor; Suravee, Sumaiya; Yordanova, Kristina
    Data annotation is a crucial step in various domains where Machine Learning (ML) approaches are utilized. Despite the availability of automated and semi-automated data labeling methods, manual annotation by experts remains essential for developing high-quality models in certain scenarios. This study explores how annotations can evolve over time through an experiment focused on annotating named entities and relationships within the domain of dementia and related behaviors. Two annotators labeled a task-specific text corpus on two separate occasions, one year apart. Our findings revealed an increase in both the quantity and quality of annotated entities and relationships for both annotators. Statistical tests were conducted to assess the significance of the changes in annotations. The results indicate substantial variability in annotations over time, particularly in complex domains. The paper also discusses potential reasons for these variations.