Logo des Repositoriums

Auflistung nach Schlagwort "semantic segmentation"

1 - 3 von 3
  • 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
    Segmenting Wood Rot using Computer Vision Models
    (INFORMATIK 2024, 2024) Kammerbauer, Roland; Schmitt, Thomas H.; Bocklet, Tobias
    In the woodworking industry, a huge amount of effort has to be invested into the initial quality assessment of the raw material. In this study, we present an AI model to detect, quantify, and localize defects on wooden logs. This model aims to both automate the quality control process and provide a more consistent and reliable quality assessment. For this purpose, a dataset of 1424 sample images of wood logs is created. A total of 5 annotators possessing different levels of expertise are involved in dataset creation. An inter-annotator agreement analysis is conducted to analyze the impact of expertise on the annotation task and to highlight subjective differences in annotator judgment. We explore, train, and fine-tune the state-of-the-art InternImage and ONE-PEACE architectures for semantic segmentation. The best model created achieves an average IoU of 0.71 and shows detection and quantification capabilities close to the human annotators.
  • Konferenzbeitrag
    Visible Wavelength Iris Segmentation: A Multi-Class Approach using Fully Convolutional Neuronal Networks
    (BIOSIG 2018 - Proceedings of the 17th International Conference of the Biometrics Special Interest Group, 2018) Osorio-Roig, Dailé; Rathgeb, Christian; Gomez-Barrero, Marta; Morales-González, Annette; Garea-Llano, Eduardo; Busch, Christoph
    Iris segmentation under visible wavelengths (VWs) is a vital processing step for iris recognition systems operating at-a-distance or in non-cooperative environments. In these scenarios the presence of various artefacts, e.g. occlusions or specular reflections, as well as out-of-focus blur represents a significant challenge. The vast majority of proposed iris segmentation algorithms under VW aim at discriminating the iris and non-iris regions without taking into account the variability that is present in the non-iris region. In this paper, we introduce the idea of segmenting the iris region using a multi-class approach which differentiates additional classes, e.g. pupil or sclera, as opposed to commonly employed bi-class approaches (iris and non-iris). Experimental results conducted on two publicly available databases show that the use of the proposed multi-class approach improves the iris segmentation accuracy. Simultaneously, it also allows for the segmentation of different non-iris regions, e.g. glasses, which could be employed in further application scenarios.