Personal details
| Title | Automatized Image Detection for Marine Macrofauna |
| Description | Location: ICBM Wilhelmshaven (& Home Office & eventually working at HIFMB) Start: As Soon as Possible Duration: 3-6 Month Background Benthic fauna communities and habitats in the North Sea are strongly impacted by anthropogenic effects such as bottom trawling. Marine Protected Areas (MPAs) were implemented to restrict anthropogenic impacts in order to protect biodiversity and ecosystem functions. A partial exclusion of bottom trawling fisheries was set in Borkum Reef and Sylt Outer Reef in 2023. A monitoring is needed to evaluate the success of MPA restrictions. Established macrofauna monitoring methods are usually destructive to the area they observe. There is a need for minimal invasive monitoring’s, like they can be done by Baited Remote Underwater Video Stations (BRUVS). This Method is being used more and more around the world and was applied in the German Bight since September 2020. By now video get analyzed manually by watching hours of video material and annotating Organisms using annotation software. Image based AI technologies developed rapidly in the last decade, becoming more user friendly and easy to train like the YOLO-Models. No advanced programming skills are needed anymore for the implementation. |
| Home institution | Institute for Chemistry and Biology of the Marine Environment |
| Associated institutions |
|
| Type of work | practical / application-focused |
| Type of thesis | Bachelor's or Master's degree |
| Author | Nicolas Bill |
| Status | available |
| Problem statement | Assignment The student's task is to work with the existing ~50000 annotations and analyze even further video material to better match the specific requirements for the creation of a powerful AI in the context of an international collaboration. AI-based results need to be compared with manually analyzed results. |
| Requirement | Requirements Good communication skills and the willingness to work in an international team are required. Programming skills in Python are useful but not mandatory. However, some programming work will need to be done during the thesis, but is nowadays easy to very easy to learn.
The applicant should have an interest in the marine macrofauna of the North Sea (taxonomic knowledge would be helpful) and not be afraid to do some programming.
The thesis can be written in English or German language.
Contact for further information: Nicolas Bill – Nicolas.bill@uni-oldenburg.de Dr. Sven Rohde – sven.rohde@uni-oldenburg.de |
| Created | 11/07/25 |