Data · dataset · 2023
Synthetic images of corals (Desmophyllum pertusum) with object detection models
Listed in Researchdata.se
Two object detection models using Darknet/YOLOv4 were trained on images of the coral Desmophyllum pertusum from the Kosterhavet National Park.
Description
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.
Read the rest (2 more)
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: github.com/NVlabs/stylegan2 YOLOv4: github.com/AlexeyAB/darknet
Links
Where it is published
- DOI doi.org/10.5878/hp35-4809 ↗
DOI / persistent id · from researchdata se
Catalogue records · 1
- OAI-PMH record api.researchdata.se/oai-pmh?verb=GetRecord&metadataPrefix=oai_dc&identifier=oai%3A… ↗
metadata API · from researchdata se
Topics
- From keywords
- Earth & Environmental Science · Life Sciences · Ocean & Atmospheric Science
Provenance · 1 source records, 11 field assertions
| Source | Key | Last seen | Raw |
|---|---|---|---|
| Researchdata.se | oai:researchdata.se:2022-98-1/1 | 11 d ago | JSON v1 |
| Field | Assertion | Extractor | Evidence |
|---|---|---|---|
| access_level | source · researchdata se | connector:researchdata_se@1.0.0 | |
| concepts[field].local:field:earth-environmental | mapping · researchdata se | connector:researchdata_se@1.0.0 | |
| concepts[field].local:field:life-sciences | mapping · researchdata se | connector:researchdata_se@1.0.0 | |
| concepts[field].local:field:ocean-atmospheric | mapping · researchdata se | connector:researchdata_se@1.0.0 | |
| concepts[modality].local:modality:image | enrichment · researchdata se | keyword-concept-rules@1.0.0 | title+description (75%) |
| concepts[modality].local:modality:video | enrichment · researchdata se | keyword-concept-rules@1.0.0 | title+description (75%) |
| description | source · researchdata se | connector:researchdata_se@1.0.0 | /metadata/dc/description |
| license | source · researchdata se | connector:researchdata_se@1.0.0 | /metadata/dc/rights |
| publication_date | source · researchdata se | connector:researchdata_se@1.0.0 | |
| spatial | source · researchdata se | connector:researchdata_se@1.0.0 | |
| title | source · researchdata se | connector:researchdata_se@1.0.0 | /metadata/dc/title |