iMagine project

Part of EU Open Research Repository
Published February 18, 2025 | Version v1
Dataset Open

Underwater images from OBSEA fish detection training dataset (YOLO)

Creators

  • 1. ROR icon Universitat Politècnica de Catalunya
  • 2. ROR icon Institut de Ciències del Mar
  • 3. Universitat Politecnica de Catalunya

Description

This dataset contains labelled underwater pictures taken at the OBSEA underwater observatory. The pictures have already been splited into train, validation and test folders for training a YOLO AI model. This is a substet of this dataset, with the data arranged to simplify the training process.

Train a YOLO model with this dataset:

  1. Install dependencies with the command  "pip3 install ultralytics"
  2. Download and unzip the file
  3. Edit the data.yaml file to match the "train", "valid" and "test" paths in your filesystem
  4. Train the model with the command: "yolo task=detect mode=train model=yolov8n.pt data=data.yaml"

Done! In order to optimize the results it is encouraged to tune the hyperparameters to fit your application. 

Training dataset

This dataset contains labeled images for fish detection acquired at OBSEA Underwater Observatory (NW Mediterranean sea).

Data preprocessing

Several data augmentation techniques have been used to improve the training. The configuration can be found in args.yaml file.

Data splitting

Data has been randomly splitted in 70% training, 20% validation and 10% test. The splits are already included in the training dataset.

Classes, labels and annotations

The following classes are included in the dataset: 

  • Chromis chromis:  WoRMS id 127000
  • Coris julis:  WoRMS id 126963
  • Dactylopterus volitans:  WoRMS id None None
  • Dentex dentex:  WoRMS id 273962
  • Diplodus cervinus:  WoRMS id 127051
  • Diplodus puntazzo:  WoRMS id 127052
  • Diplodus sargus:  WoRMS id 127053
  • Diplodus vulgaris:  WoRMS id 127054
  • Diver:  scuba diver, used mainly to prevent divers to be detected as fish
  • Epinephelus costae:  WoRMS id 127034
  • Epinephelus marginatus:  WoRMS id 127036
  • Mullus surmuletus:  WoRMS id 126986
  • Muraena helena:  WoRMS id 126303
  • Myliobatidae:  WoRMS id None None
  • Oblada melanura:  WoRMS id None None
  •  Octopus vulgaris:  WoRMS id None None
  •  Parablennius gattorugine:  WoRMS id 126770
  •  Sarpa salpa:  WoRMS id 127064
  •  Sciaena umbra:  WoRMS id None None
  •  Seriola dumerili:  WoRMS id 126816
  •  Serranus cabrilla:  WoRMS id 127041
  •  Sparus aurata:  WoRMS id 151523
  •  Symphodus mediterraneus:  WoRMS id 273569

Data sources

Pictures where acquired by several underwater cameras, deployed at OBSEA, model Linovision IPC608.

Data quality

Images have been manually selected to include as much variety as possible in terms of light and water turbidity.

Spatial coverage

All pictures where taken at OBSEA underwater observatory, off-the-coast of Vilanova i la Geltrú, Spain. GPS coordinates

 Longitude   Latitude depth
1.75257 41.18212 20 m

Contact information

For further technical inquiries or additional information about the annotated dataset, please contact enoc.martinez@upc.edu

 

Files

obsea_split_YOLO.zip

Files (3.8 GB)

Name Size Download all
md5:80daf33935f424e66fb35e31ef3b1780
3.8 GB Preview Download