Deep learning-based localisation of combine harvester components in thermal images
H. Senke, D. Sprute, U. Büker, H. Flatt, Deep Learning-Based Localisation of Combine Harvester Components in Thermal Images, KIT Scientific Publishing, Karlsruhe, 2024.
Download
Es wurde kein Volltext hochgeladen. Nur Publikationsnachweis!
Konferenzband - Beitrag
| Veröffentlicht
| Englisch
Autor*in
Herausgeber*in
Längle, Thomas;
Heizmann, Michael
Körperschaftlicher Herausgeber
Karlsruher Institut für Technologie. Institut für Industrielle Informationstechnik ;
Fraunhofer-Institut für Optronik, Systemtechnik und Bildauswertung
Abstract
It is crucial to identify defective machine components in production to ensure quality. Some components generate heat when defective, so automating the inspection process with a thermal imaging camera can provide qualitative measurements. This work aims to use computer vision methods to locate these components in thermal images. Since there is currently no comparison of object detection and semantic segmentation algorithms for this use case, this study compares different architectures with the goal of localising these components for further defect inspection. Moreover, as there are currently no datasets for this use case, this study contributes a novel annotated dataset of thermal images of combine harvester components. The different algorithms are evaluated based on the quality of their predictions and their suitability for further defect inspection. As semantic segmentation and object detection cannot be directly compared with each other, custom weighted metrics are used. The architectures evaluated include RetinaNet, YOLOV8 Detector, DeepLabV3+, and SegFormer. Based on the experimental results, semantic segmentation outperforms object detection regarding the use case, and the SegFormer architecture achieves the best results with a weighted MeanIOU of 0.853.
Stichworte
Erscheinungsjahr
Titel Konferenzband
Forum Bildverarbeitung 2024 = Image Pocessing Forum 2024
Seite
71-82
Konferenz
Forum Bildverarbeitung 2024
Konferenzort
Karlsruhe
Konferenzdatum
2024-11-21 – 2024-11-22
ISBN
ELSA-ID
Zitieren
Senke H, Sprute D, Büker U, Flatt H. Deep Learning-Based Localisation of Combine Harvester Components in Thermal Images. (Längle T, Heizmann M, Karlsruher Institut für Technologie. Institut für Industrielle Informationstechnik , Fraunhofer-Institut für Optronik, Systemtechnik und Bildauswertung , eds.). KIT Scientific Publishing; 2024:71-82. doi:10.58895/ksp/1000174496-7
Senke, H., Sprute, D., Büker, U., & Flatt, H. (2024). Deep learning-based localisation of combine harvester components in thermal images. In T. Längle, M. Heizmann, Karlsruher Institut für Technologie. Institut für Industrielle Informationstechnik , & Fraunhofer-Institut für Optronik, Systemtechnik und Bildauswertung (Eds.), Forum Bildverarbeitung 2024 = Image Pocessing Forum 2024 (pp. 71–82). KIT Scientific Publishing. https://doi.org/10.58895/ksp/1000174496-7
Senke H et al. (2024) Deep Learning-Based Localisation of Combine Harvester Components in Thermal Images, Längle T et al. (eds). Karlsruhe: KIT Scientific Publishing.
Senke, Hanna, Dennis Sprute, Ulrich Büker, and Holger Flatt. Deep Learning-Based Localisation of Combine Harvester Components in Thermal Images. Edited by Thomas Längle, Michael Heizmann, Karlsruher Institut für Technologie. Institut für Industrielle Informationstechnik , and Fraunhofer-Institut für Optronik, Systemtechnik und Bildauswertung . Forum Bildverarbeitung 2024 = Image Pocessing Forum 2024. Karlsruhe: KIT Scientific Publishing, 2024. https://doi.org/10.58895/ksp/1000174496-7.
Senke, Hanna, Dennis Sprute, Ulrich Büker und Holger Flatt. 2024. Deep learning-based localisation of combine harvester components in thermal images. Hg. von Thomas Längle, Michael Heizmann, Karlsruher Institut für Technologie. Institut für Industrielle Informationstechnik , und Fraunhofer-Institut für Optronik, Systemtechnik und Bildauswertung . Forum Bildverarbeitung 2024 = Image Pocessing Forum 2024. Karlsruhe: KIT Scientific Publishing. doi:10.58895/ksp/1000174496-7, .
Senke, Hanna ; Sprute, Dennis ; Büker, Ulrich ; Flatt, Holger ; Längle, T. ; Heizmann, M. ; Karlsruher Institut für Technologie. Institut für Industrielle Informationstechnik ; Fraunhofer-Institut für Optronik, Systemtechnik und Bildauswertung (Hrsg.): Deep learning-based localisation of combine harvester components in thermal images. Karlsruhe : KIT Scientific Publishing, 2024
H. Senke, D. Sprute, U. Büker, H. Flatt, Deep learning-based localisation of combine harvester components in thermal images, KIT Scientific Publishing, Karlsruhe, 2024.
H. Senke, D. Sprute, U. Büker, and H. Flatt, Deep learning-based localisation of combine harvester components in thermal images. Karlsruhe: KIT Scientific Publishing, 2024, pp. 71–82. doi: 10.58895/ksp/1000174496-7.
Senke, Hanna, et al. “Deep Learning-Based Localisation of Combine Harvester Components in Thermal Images.” Forum Bildverarbeitung 2024 = Image Pocessing Forum 2024, edited by Thomas Längle et al., KIT Scientific Publishing, 2024, pp. 71–82, https://doi.org/10.58895/ksp/1000174496-7.
Senke, Hanna u. a.: Deep learning-based localisation of combine harvester components in thermal images, hg. von Längle, Thomas u. a., Karlsruhe 2024.
Senke H, Sprute D, Büker U, Flatt H. Deep learning-based localisation of combine harvester components in thermal images. Längle T, Heizmann M, Karlsruher Institut für Technologie. Institut für Industrielle Informationstechnik , Fraunhofer-Institut für Optronik, Systemtechnik und Bildauswertung , editors. Forum Bildverarbeitung 2024 = Image Pocessing Forum 2024. Karlsruhe: KIT Scientific Publishing; 2024.