Otwarty dostęp

Seafloor mapping based on multibeam echosounder bathymetry and backscatter data using Object-Based Image Analysis: a case study from the Rewal site, the Southern Baltic

Łukasz Janowski
Łukasz Janowski
, 
Jarosław Tęgowski
Jarosław Tęgowski
 oraz 
Jarosław Nowak
Jarosław Nowak
   | 21 wrz 2018
Oceanological and Hydrobiological Studies's Cover Image
O artykule
Poprzedni artykuł
Następny artykuł
Zacytuj
Article Category: Original research paper
Data publikacji: 21 wrz 2018
Zakres stron: 248 - 259
Otrzymano: 04 paź 2017
Przyjęty: 16 sty 2018
DOI: https://doi.org/10.1515/ohs-2018-0024
© 2018 Faculty of Oceanography and Geography, University of Gdańsk, Poland
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1

Location of the Rewal study site within the Polish EEZ. Sources: our study, OpenStreetMap, European Environment Agency
Location of the Rewal study site within the Polish EEZ. Sources: our study, OpenStreetMap, European Environment Agency

Figure 2

Distribution of the mean backscatter intensity in four ground-truth classes. Sources: our study
Distribution of the mean backscatter intensity in four ground-truth classes. Sources: our study

Figure 3

A) result of bathymetric processing; B) result of backscatter processing and location of ground-truth samples, division into training and validation ground-truth samples; C) result of rugosity extracted as a secondary derivative of bathymetry; D) result of CART classification based on Object-Based Image Analysis. Sources: our study
A) result of bathymetric processing; B) result of backscatter processing and location of ground-truth samples, division into training and validation ground-truth samples; C) result of rugosity extracted as a secondary derivative of bathymetry; D) result of CART classification based on Object-Based Image Analysis. Sources: our study

Figure 4

CART results. Sources: our study
CART results. Sources: our study

Error matrix of the CART classification based on the OBIA technique. MS – Muddy Sand, C – Clay, FS – Fine Sand, VFS – Very Fine Sand. Sources: our study

Refererce Class
User MS C FS VFS Sum
MS 2 0 0 1 3
C 0 0 0 0 0
FS 0 0 2 0 2
VFS 0 0 0 3 3
Sum 2 0 2 4
Producer 1 undefined 1 0.75
User 0.666667 undefined 1 1
Overall Accuracy 0.875
KIA 0.809524

List of all ground-truth samples with their descriptions and locations. Sources: our study

ID Sediment type Class Type Latitude Longitude
VR 201 Very Fine Sand VFS training 54.105263 14.911653
VR 202 Fine Sand FS training 54.108030 14.911664
VR 203 Muddy Sand MS training 54.120947 14.911780
VR 204 Muddy Sand MS validation 54.127028 14.911643
VR 205 Very Fine Sand VFS training 54.111783 14.926790
VR 206 Very Fine Sand VFS training 54.114840 14.926752
VR 207 Fine Sand FS training 54.126588 14.926762
VR 208 Very Fine Sand VFS validation 54.120305 14.942225
VR 209 Fine Sand FS validation 54.110749 14.942321
VR 210 Very Fine Sand VFS validation 54.106412 14.972621
VR 211 Very Fine Sand VFS validation 54.122082 14.972553
VR 212 Muddy Sand MS validation 54.130222 14.972554
VR 214 Fine Sand FS validation 54.116665 14.987440
VR 215 Very Fine Sand VFS training 54.111829 14.987819
VR 216 Very Fine Sand VFS validation 54.109825 14.987764
VR 225 Muddy Sand MS training 54.125345 14.896156
VR 226 Very Fine Sand VFS training 54.118745 14.896162
VR 227 Clay C training 54.109162 14.896139
VR 228 Muddy Sand MS training 54.106042 14.896208
eISSN:
1897-3191
Język:
Angielski
Częstotliwość wydawania:
4 razy w roku
Dziedziny czasopisma:
Chemistry, other, Geosciences, Life Sciences
Kanał RSS czasopisma