IMR_2025_Meso_DeepVision

Annotated In-Trawl Camera Images of Mesopelagic Organisms: A Dataset for Ecological and Machine Learning Applications

Dataset Creator	Dataset Title	Dataset Release Date	Dataset Publisher	Dataset DOI	Online Resource
Vaneeda Allken, Taraneh Westergerling, Webjørn Melle	In-trawl camera images and labels of mesopelagic organisms.	2025-11-04T00:00:00:00Z	Institute of Marine Research	10.21335/NMDC-1444357535	http://metadata.nmdc.no/metadata-api/landingpage/5d9089712c24ec6d810e50ea6b7d9655

Role	First Name	Last Name	Email
Technical Contact	Webjørn	Melle	webjoern.melle@hi.no
DIF Author	Arnfinn	Morvik	arnfinn.morvik@hi.no
Investigator	Vaneeda	Allken	vaneeda.allken@hi.no
Technical Contact	Taraneh	Westergerling

Category	Topic	Term	Variable Level 1	Variable Level 2
EARTH SCIENCE	BIOLOGICAL CLASSIFICATION	ANIMALS/INVERTEBRATES	ARTHROPODS	CRUSTACEANS
EARTH SCIENCE	BIOLOGICAL CLASSIFICATION	ANIMALS/INVERTEBRATES	CNIDARIANS	JELLYFISHES
EARTH SCIENCE	BIOLOGICAL CLASSIFICATION	ANIMALS/VERTEBRATES	FISH
EARTH SCIENCE	OCEANS	AQUATIC SCIENCES	FISHERIES
EARTH SCIENCE SERVICES	MACHINE LEARNING TRAINING DATA	SOURCE	RASTER SOURCE
EARTH SCIENCE SERVICES	MACHINE LEARNING TRAINING DATA	LABELS	VECTOR LABEL
EARTH SCIENCE SERVICES	MODELS	MACHINE LEARNING MODELS	CLASSIFICATION

Oceans

deep vision,image analysis,mesopelagic,object detection,trawl video,trawl sampling,YOLO,barracudina,fish,gelatinous,krill,lanternfish,pelagic shrimp,silvery lightfish,squid

Start Date	Stop Date
2020-05-24	2024-03-02

Complete

Location Category	Location Type	Location Subregion1	Location Subregion2
Ocean	Atlantic Ocean	North Atlantic Ocean	Norwegian Sea

unrestricted

Institute of Marine Research

Data Center Name	Personnel	Data Center URL
Institute of Marine Research, Norwegian Marine Data Center Institute of Marine Research, Norwegian Marine Data Center	Datacenter Arnfinn Morvik arnfinn.morvik@hi.no
NO/IMR Institute of Marine Research	Originating institution Webjørn Melle webjoern.melle@hi.no	http://www.imr.no

Distribution Media	Distribution Format
ftp	zip

This repository provides a structured collection of datasets and configuration files used for training and evaluating object detection models to detect mesopelagic organisms from images. The datasets (described in the readme file) are stored in CSV format under datasets/csv_files, where each file contains image paths, bounding box coordinates, species class labels, and image names. Multiple dataset variants (described in the manuscript) are included, covering different colour corrections (e.g., red CC, red non-CC, white), augmentation strategies, and splits (train, validation, test). For YOLO-based training, the data is organized under datasets/YOLO, with separate folders for image paths (images/), YOLO-format labels (labels/), and dataset configuration files (.yaml). The deepvision/ directory contains annotated data, separated into rectified and non-rectified formats, along with original XML annotations and session-level metadata in meta_per_deep_vision_session.csv as well. It also contains synthetic data used in some dataset configurations for training, under /train. The model/ directory contains an example training script and a configuration file with the parameters used for training. This structure supports flexible experimentation with various data configurations and model training pipelines, facilitating reproducibility and scalability in object detection tasks.

URL Content Type	URL
GET DATA	https://ftp.nmdc.no/nmdc/IMR/mesopelagic/datasets.zip
GET DATA	https://ftp.nmdc.no/nmdc/IMR/mesopelagic/model.zip
GET DATA	https://ftp.nmdc.no/nmdc/IMR/mesopelagic/labelled.zip
GET DATA	https://ftp.nmdc.no/nmdc/IMR/mesopelagic/train.zip
VIEW RELATED INFORMATION	https://ftp.nmdc.no/nmdc/IMR/mesopelagic/readme.docx

Norwegian Marine Datacentre

CEOS IDN DIF

9.7.1

2025-11-04T00:00:00:00Z

CC BY 4.0

POLYGON((20.299999999999997 60,-11.3 60,-11.3 73.69999999999999,20.299999999999997 73.69999999999999,20.299999999999997 60))