DPLUS0045 Anguilla Seabed Classification derived from Pleiades data - T2
Seabed classification and habitat map created from remote sensing data as part of the Darwin Plus funded project "Mapping Anguilla’s ‘Blue Belt’ Ecosystem Services". The project develops local capacity to undertake comprehensive marine resource assessments of Anguilla’s ‘Blue-Belt’, delivering best practice guidance for combining satellite and acoustic surveys, to extend mapping into deeper waters for the first time. The complexity of this shapefile means that it can only be dowloaded as a binary file and is not available via WMS/WFS.
dataset
https://data.cefas.co.uk/view/19318
name: Cefas Data Portal
description: The Cefas Data Portal contains metadata records and data sets available to download and connect to in support of our commitment to open science. Data is available in the following formats: Binary download.
function: download
CEFAS19318
https://data.cefas.co.uk
eng
biota
publication
2008-06-01
revision
2011-03-25
revision
2011-03-25
publication
2008-06-01
revision
2010-05-18
publication
2012-01-11
publication
2012-01-11
-70.00
-60.00
20.00
10.00
2016-09-04
2018-03-30
publication
2019-02-01
revision
2023-02-06
creation
2018-08-29
notPlanned
Seabed classification and habitat maps were created from the remote sensing data collected by the Pleiades 1A Sensor. The data was mapped using Object based image analysis, a two-step process which involves the segmentation and then classification of an image. Before segmentation and classification, the satellite data underwent several different corrections including radiometric calibration, atmospheric correction, orthorectification, land and cloud masking, sun glint correction and depth correction. The satellite data was segmented using the multi-resolution segmentation algorithm within Trimble’s eCognition® software (v9.0.0). The objects produced during the segmentation were classified using a mixture of K-nearest neighbour, random forest and support vector machines classification algorithms. All three of these algorithms take the statistics from the training dataset, created from the ground truthing data, and classify the remaining objects based on statistically driven decisions. Data was classified to the more comprehensive habitat descriptors. Maps were validated against a subset of the groundtruthing stations which were left out of the classification. The map achieved an overall accuracy of 52.9%.
publication
2013-12-10
false
See the referenced specification
publication
2010-12-08
true
See the referenced specification
Unknown
Centre for Environment, Fisheries and Aquaculture Science, Lowestoft Laboratory (CEFAS)
Cefas Lowestoft Laboratory
Pakefield Road
Lowestoft
NR33 0HT
UK
originator
Centre for Environment, Fisheries and Aquaculture Science, Lowestoft Laboratory (CEFAS)
Cefas Lowestoft Laboratory
Pakefield Road
Lowestoft
NR33 0HT
UK
custodian
Centre for Environment, Fisheries and Aquaculture Science, Lowestoft Laboratory (CEFAS)
Cefas Lowestoft Laboratory
Pakefield Road
Lowestoft
NR33 0HT
UK
distributor
Centre for Environment, Fisheries and Aquaculture Science, Lowestoft Laboratory (CEFAS)
Cefas Lowestoft Laboratory
Pakefield Road
Lowestoft
NR33 0HT
UK
pointOfContact
2023-02-06T01:22:10