Paving the way from in situ plankton image data to a Digital Twin Ocean
Acronym
DIPLO
Titel
Paving the way from in situ plankton image data to a Digital Twin Ocean
Kurzbeschreibung
The field of plankton imaging is rapidly growing. So far, a variety of data processing tools are used for each instrument, specifically tailored to its optical properties. This lack of a consistent cross-platform protocol for data analysis impedes the inter-comparability of instruments, and integration into a harmonized global dataset. Here, DIPLO will develop a novel, efficient and user-friendly data processing pipeline to analyze image data independent of the instrument with which raw data was collected, covering the entire data processing task from raw images to ecologically meaningful data, e.g. segmentation of raw images and classification.
This project will combine a suite of modern approaches from computer vision, artificial intelligence and deep learning (e.g. convolutional neural networks). The result will be a consistent and harmonized cross-platform 4-D plankton image database, which is a critical step towards an image-based ecosystem component of a “Digital Twin Ocean”.
Start
Januar 2023
Ende
Dezember 2025
Bewilligungssumme (gesamt)
-
Bewilligungssumme (GEOMAR)
2000000
Zuwendungsgeber / Programm
-
Helmholtz-Gemeinschaft
/ Helmholtz Imaging
Koordination
Helmholtz-Zentrum für Ozeanforschung Kiel (GEOMAR), Germany
Kontakt
Partner
Alfred-Wegener-Institute (AWI), Germany
