https://researchdata.se/en/catalogue Search results en Two object detection models using Darknet/YOLOv4 were trained on images of the coral Desmophyllum pertusum from the Kosterhavet National Park. In one of the models, the training image data was amplified using StyleGAN2 generative modeling. The dataset contains 2266 synthetic images with labels and 409 original images of corals used for training the ML model. Included is also the YOLOv4 models and the StyleGAN2 network. The still images were extracted from raw video data collected using a remotely operated underwater vehicle. 409 JPEG images from the raw video data are provided in 720x576 resolution. In certain images, coordinates visible in the OSD have been cropped. The synthetic images are PNG files in 512x512 resolution. The StyleGAN2 network is included as a serialized pickle file (*.pkl). The object detection models are provided in the .weights format used by the Darknet/YOLOv4 package. Two files are included (trained on original images only, trained on original + synthetic images). The machine learning software packages used is currently (2022) available on Github: StyleGAN2: https://github.com/NVlabs/stylegan2 YOLOv4: https://github.com/AlexeyAB/darknet Wed, 12 Apr 2023 09:35:40 GMT https://researchdata.se/en/catalogue/dataset/2022-98-1 https://researchdata.se/en/catalogue/dataset/2022-98-1 University of Gothenburg Matthias Obst Sarah Al-Khateeb Victor Anton Jannes Germishuys The datasets are based on laboratory studies with larvae of the cold-water scleractinian coral Desmophyllum pertusum (syn. Lophelia pertusa) testing larval responses to nine different substrates. The experiment was run in standard micro well plates (6 wells per plate, 34 mm ø), with substrates pressed into a coral epoxy putty to level the surface. Larvae were filmed for 5 minutes as they were added to the wells. There were 3 replicate wells per substrate type and a total of 356 larvae (11–15 per well). The videos were then analyzed in the BORIS software where 8 defined behaviors were annotated to get time spent on each of the behaviors. We also counted the number of attached larvae in the wells. Dataset publication was supported by Mistra Co-Creating Better Blue (C2B2). Mon, 15 Sep 2025 12:12:47 GMT https://researchdata.se/en/catalogue/dataset/2025-43 https://researchdata.se/en/catalogue/dataset/2025-43 University of Gothenburg Susanna M. Strömberg Ann I. Larsson The datasets are based on data collected from field trials with settling panels, testing the effect of different concrete mixes on recruitment of benthic fauna. There were three different concrete mixes based on standard Portland cement and ground granulated blast furnace slag (GGBFS) and silica oxide of different ratios, compared to panels made of standard Portland cement. The factors investigated were mix (4 levels), surface (2 levels), side (4 levels) and row (4 levels). The main factors were mix and surface, relating to material composition and level of complexity. Side and row was added factors relating to how the panels were mounted on the square rack, and were affected by local hydrodynamics and boundary layer effects. Dataset publication was supported by Mistra Co-Creating Better Blue (C2B2). Mon, 13 Oct 2025 12:57:02 GMT https://researchdata.se/en/catalogue/dataset/2025-50 https://researchdata.se/en/catalogue/dataset/2025-50 University of Gothenburg Susanna M. Strömberg Ann I. Larsson Supplementary tables for the Review article "Connectivity and Population Structure in a Marginal Sea - a Review" (https://doi.org/10.1111/ddi.70056). Contains information on connectivity and population structure within marine species inhabiting the Skagerrak sea, summarised from scientific studies published between the years 1990 and 2023. Table S1: List of all 484 scientific studies that were screened, as well as information regarding whether they were included in the literature review. Table S2: Contains extracted data from the 172 scientific studies, published between 1990 and 2023, that were reviewed with regard to connectivity and population structure in the Skagerrak. The data includes information about both the studies' methodology (such as scientific discipline, spatial and temporal scale, and sample size) and results (such as which populations were identified and estimated connectivity between different regions). Table S3: Detailed summary of the scientific studies that studied contemporary connectivity on a "large" scale (between the Skagerrak sea and adjacent seas). For species with information from several studies, a concensus result for the species was created. Table S4: Detailed summary of the scientific studies that studied contemporary connectivity on a "small" scale (within the Skagerrak sea). For species with information from several studies, a concensus result for the species was created. Table S5: Detailed summary of the scientific studies that studied population structure, both within the Skagerrak Sea and between the Skagerrak and adjacent seas. For species with information from several studies, a concensus result for the species was created. Tables S6-S7: Settings for the BARRIER analysis conducted in the study. For more detailed descriptions of the content we refer to the text in the original study. This dataset is listed as "Deliverable 2.2" of the MARHAB project (Horizon Europe, grant number 101135307). Mon, 10 Nov 2025 12:52:33 GMT https://researchdata.se/en/catalogue/dataset/2025-294 https://researchdata.se/en/catalogue/dataset/2025-294 University of Gothenburg Simon Henriksson Per Erik Jorde Charlotte Berkström Guldborg Søvik Pierre De Wit Halvor Knutsen Even Moland Carl André Marlene Jahnke