Published July 16, 2024 | Version 1
Dataset Open

Nephrops (Nephrops norvegicus) Burrow object detection simple training dataset from Irish Underwater TV surveys

  • 1. Student

Contributors

  • 1. ROR icon Marine Institute

Description

Training dataset

Norway prawns (Nephrops norvegicus), also known as the Dublin Bay prawn, are common around the Irish coast. They are found in distinct sandy/muddy areas where the sediment is suitable for them to construct their burrows. Nephrops spend a great deal of time in their burrows and their emergence from these is related to time of year, light intensity and tidal strength. The Irish Nephrops fishery is extremely valuable with landings recently worth around €55m at first sale, supporting an important Irish fishing industry. 

Nephrops are managed in Functional Units (FUs). The Marine Institute has conducted under water television surveys since 2002 to independently estimate abundance, distribution and stock sizes of Nephrops norvegicus for:

  • Irish Sea Nephrops Grounds (FU 14 and 15) in collaboration with AFBI an CEFAS.
  • Porcupine Bank Nephrops Grounds (FU16)
  • Aran, Galway Bay and Slyne Head Nephrops Grounds (FU17)
  • South and South west Ireland Nephrops Grounds (FU19)
  • Labadie, Jones and Cockburn Nephrops Grounds (FU20 and 21)
  • “Smalls” Nephrops Grounds (FU22)

Each year during the summer months, on average 300 stations are surveyed each year, in three survey legs, covering all the FUs in depths from 20 to 650 metres.

A high definition camera system is towed over the sea bed for 10 minutes travelling approx. 200m at 0.8 knots on a purpose built sledge. The UWTV survey follows survey protocols available here agreed by International Council for the Exploration of the Sea (ICES) Working Group on Nephrops surveys (WGNEPS). 

As part of the iMagine project a selection of images from the Underwater TV survey Functional Units were annotated with bounding boxes and labels in YOLOv8 format to train an YOLOv8 Object Detection Models. The training dataset is saved in YOLOv8 format.  It is intended to train a YOLOv8 Nephrrops burrow object detection model to assess the utility of an Object Detection model is assisting Prawn Survey work in the semi automated annotation of prawn burrow imagery.

Technical info (En)


Data preprocessing

There was no data preprocessing steps applied, However data was augmented using the Roboflow platform. 

The following augmentation was applied to create 3 versions of each source training image:
* Random exposure adjustment of between -13 and +13 percent
* Salt and pepper noise was applied to 0.26 percent of pixels

Data splitting

The initial training dataset was split as follows: 90% Training Set, 6%  Validation Set and 4% Test Set.

Data labelling

closed_burrow, crab_burrow, gatekept_burrow, null, prawn_burrow, small_prawn_burrow

Parameters

NA

Data sources

A high definition camera system is towed over the sea bed for 10 minutes travelling approx. 200m at 0.8 knots on a purpose built sledge. The UWTV survey follows survey protocols available here agreed by International Council for the Exploration of the Sea (ICES) Working Group on Nephrops surveys (WGNEPS). 

Data Quality

The sample of Images used was taken from the Nephrops Survey data archive.

Data resizing

The images are not resized. During the yolov8 training process the images are resized to 640x640 by default.

Spatial Coverage

ICES Working Group for the Celtic Seas Ecoregion (WGCSE) Nephrops Functional Units(FU):

Irish Sea Nephrops Grounds (FU 14 and 15) 

Porcupine Bank Nephrops Grounds (FU16)

Aran, Galway Bay and Slyne Head Nephrops Grounds (FU17)

South and South west Ireland Nephrops Grounds (FU19)

Labadie, Jones and Cockburn Nephrops Grounds (FU20 and 21)

“Smalls” Nephrops Grounds (FU22)

Contact information

Data.Requests@Marine.ie

Files

nephrops_burrows_simple.v1i.yolov8.zip

Files (66.5 MB)

Name Size Download all
md5:abf3c39c7ce8503f3aa3cd9038917ab1
66.5 MB Preview Download

Additional details

Funding

European Commission
iMagine - Imaging data and services for aquatic science 101058625

Dates

Created
2024-07-16
Dataset created in yolov8 format in Roboflow
Collected
2019-06-14
Images collected from
Collected
2023-06-20
Images collected to

Biodiversity