Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023;140(6):519-529.
doi: 10.1007/s00502-023-01158-w. Epub 2023 Sep 12.

[Recognizing transparent objects for laboratory automation]

[Article in German]
Markus Vincze  1 , Jean-Baptiste Weibel  1 , Stefan Thalhammer  1 , Hrishikesh Gupta  1 , Philipp Ausserlechner  1
Affiliations

[Recognizing transparent objects for laboratory automation]

[Article in German]
Markus Vincze et al. Elektrotech Informationstechnik. 2023.

Abstract
in English, German

While matte objects can be visually recognized well and grasped with robots, transparent objects pose new challenges. Modern color and depth cameras (RGB-D) do not deliver correct depth data but distorted images of the background. In this paper, we show which methods are suitable to detect transparent objects in color images only and to determine their pose. Using a robotic system, views of the targeted object are generated and annotated to learn methods and to obtain data for evaluation. We also show that by using an improved method for fitting the 3D pose, a significant improvement in the accuracy of pose estimation is achieved. Thus, false detections can be eliminated and for correct detections the accuracy of pose estimation is improved. This makes it possible to grasp transparent objects with a robot.

Während matte Objekte visuell gut erkannt und mit Robotern gegriffen werden können, stellen transparente Objekte neue Herausforderungen dar. So liefern moderne Farb- und Tiefenbildkameras (RGB-D) keine korrekten Tiefendaten, sondern verzerrte Abbildungen des Hintergrunds. Wir zeigen in diesem Beitrag, welche Methoden geeignet sind, um nur in Farbbildern transparente Objekte zu erkennen und deren Pose zu bestimmen. Mittels eines Robotersystems werden Ansichten des Zielobjekts generiert und annotiert, um Methoden anzulernen und um Daten für die Evaluierung zu erhalten. Wir zeigen auch, dass mittels einer verbesserten Methode zum Einpassen der 3D-Pose eine deutliche Verbesserung der Genauigkeit der Lageschätzung erreicht wird. Dadurch können falsche Erkennungen aussortiert werden und für richtige Erkennungen wird die Genauigkeit der Poseschätzung verbessert. Somit gelingt es, mit einem Roboter transparente Objekte zu greifen.

Keywords: Computer vision; Detection; Grasping; Pose estimation; Robotics; Transparent objects.

PubMed Disclaimer

References

    1. Biermann, F., Mathews, J., Nießing, B., König, N., Schmitt, R.H.: Automating laboratory processes by connecting biotech and robotic devices—an overview of the current challenges, existing solutions and ongoing developments. Processes 9(6), 966 (2021). 10.3390/pr9060966
    1. Charoo, N.A., Khan, M.A., Rahman, Z.: Data integrity issues in pharmaceutical industry: Common observations, challenges and mitigations strategies. International Journal of Pharmaceutics 631, 122503 (2023). 10.1016/j.ijpharm.2022.122503 - PubMed
    1. Vagaš, M., Putala, J.: Verification of designed assembly process at research robotized workplace. Applied Mechanics and Materials 844, 38–43 (2016). 10.4028/www.scientific.net/amm.844.38
    1. Rathmair, M., Haspl, T., Komenda, T., Reiterer, B., Hofbaur, M.: A formal verification approach for robotic workflows. In: 2021 IEEE 20th International Conference on Advanced Robotics (ICAR), pp. 670–675 (2021). 10.1109/icar53236.2021.9659366
    1. Rojas, J., Harada, K., Onda, H., Yamanobe, N., Yoshida, E., Nagata, K., Kawai, Y.: Probabilistic state verification for snap assemblies using the relative-change-based hierarchical taxonomy. In: 2012 12th IEEE-RAS International Conference on Humanoid Robots (Humanoids 2012) (2012). 10.1109/humanoids.2012.6651505

Publication types

LinkOut - more resources