Abstract
necessary to understand which of the commonly used segmentation methods is best suited to account for such level of detail. At the same time, new sampling protocols for precisely georeferenced ground truth data need to be developed to validate the benthic environmental classification. This study focuses on a dataset collected in a shallow (2–10 m deep) tidal channel of the Lagoon of Venice, Italy. Using 0.05-m and 0.2-m raster grids, we compared a range of classifications, both pixel- based and object-based approaches, including manual, Maximum Likelihood Classifier, Jenks Optimization clustering, textural analysis and Object
Based Image Analysis. Through a comprehensive and accurately geo-referenced ground truth dataset, we were able to identify five different classes of the substrate composition, including sponges, mixed submerged aquatic vegetation, mixed detritic bottom (fine and coarse) and unconsolidated bare sediment. We computed estimates of accuracy (namely Overall, User and Producer Accuracies) by cross tabulating predicted and reference instances. Overall, pixel based segmentations produced the highest accuracies and that the accuracy assessment is strongly dependent on the choice of classes for the segmentation. Tidal channels in the Venice Lagoon are extremely important in terms of habitats and sediment distribution, particularly within the context of the new tidal barrier being built. However, they had remained largely unexplored until now, because of the surveying challenges. The application of this remote sensing approach, combined with targeted sampling, opens a new perspective in the monitoring of benthic habitats in view of a knowledge-based management of natural resources in shallow coastal areas.
| Original language | English |
|---|---|
| Pages (from-to) | 45-60 |
| Number of pages | 16 |
| Journal | Estuarine, Coastal and Shelf Science |
| Volume | 170 |
| Early online date | 17 Dec 2015 |
| DOIs | |
| Publication status | Published - 5 Mar 2016 |
Fingerprint
Keywords
- Benthic habitat mapping
- high-resolution sonar
- image segmentation
- very shallow water
- multibeam
- Venice Lagoon
Cite this
Evaluation of seabed mapping methods for fine-scale classification of extremely shallow benthic habitats – application to the Venice Lagoon, Italy. / Montereale Gavazzi, Giacomo; Madricardo, Fantina; Janowski, Lukas; Kruss, Aleksandra; Blondel, Philippe; Sigovini, Marco; Foglini, Federica.
In: Estuarine, Coastal and Shelf Science, Vol. 170, 05.03.2016, p. 45-60.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Evaluation of seabed mapping methods for fine-scale classification of extremely shallow benthic habitats – application to the Venice Lagoon, Italy
AU - Montereale Gavazzi, Giacomo
AU - Madricardo, Fantina
AU - Janowski, Lukas
AU - Kruss, Aleksandra
AU - Blondel, Philippe
AU - Sigovini, Marco
AU - Foglini, Federica
PY - 2016/3/5
Y1 - 2016/3/5
N2 - Recent technological developments of multibeam echosounder systems (MBES) allow mapping of benthic habitats with unprecedented detail. MBES can now be employed in extremely shallow waters, challenging data acquisition (as these instruments were often designed for deeper waters) and data interpretation (honed on datasets with resolution sometimes orders of magnitude lower). With extremely high-resolution bathymetry and colocated backscatter data, it is now possible to map the spatial distribution of fine scale benthic habitats, even identifying the acoustic signatures of single sponges. In this context, it isnecessary to understand which of the commonly used segmentation methods is best suited to account for such level of detail. At the same time, new sampling protocols for precisely georeferenced ground truth data need to be developed to validate the benthic environmental classification. This study focuses on a dataset collected in a shallow (2–10 m deep) tidal channel of the Lagoon of Venice, Italy. Using 0.05-m and 0.2-m raster grids, we compared a range of classifications, both pixel- based and object-based approaches, including manual, Maximum Likelihood Classifier, Jenks Optimization clustering, textural analysis and ObjectBased Image Analysis. Through a comprehensive and accurately geo-referenced ground truth dataset, we were able to identify five different classes of the substrate composition, including sponges, mixed submerged aquatic vegetation, mixed detritic bottom (fine and coarse) and unconsolidated bare sediment. We computed estimates of accuracy (namely Overall, User and Producer Accuracies) by cross tabulating predicted and reference instances. Overall, pixel based segmentations produced the highest accuracies and that the accuracy assessment is strongly dependent on the choice of classes for the segmentation. Tidal channels in the Venice Lagoon are extremely important in terms of habitats and sediment distribution, particularly within the context of the new tidal barrier being built. However, they had remained largely unexplored until now, because of the surveying challenges. The application of this remote sensing approach, combined with targeted sampling, opens a new perspective in the monitoring of benthic habitats in view of a knowledge-based management of natural resources in shallow coastal areas.
AB - Recent technological developments of multibeam echosounder systems (MBES) allow mapping of benthic habitats with unprecedented detail. MBES can now be employed in extremely shallow waters, challenging data acquisition (as these instruments were often designed for deeper waters) and data interpretation (honed on datasets with resolution sometimes orders of magnitude lower). With extremely high-resolution bathymetry and colocated backscatter data, it is now possible to map the spatial distribution of fine scale benthic habitats, even identifying the acoustic signatures of single sponges. In this context, it isnecessary to understand which of the commonly used segmentation methods is best suited to account for such level of detail. At the same time, new sampling protocols for precisely georeferenced ground truth data need to be developed to validate the benthic environmental classification. This study focuses on a dataset collected in a shallow (2–10 m deep) tidal channel of the Lagoon of Venice, Italy. Using 0.05-m and 0.2-m raster grids, we compared a range of classifications, both pixel- based and object-based approaches, including manual, Maximum Likelihood Classifier, Jenks Optimization clustering, textural analysis and ObjectBased Image Analysis. Through a comprehensive and accurately geo-referenced ground truth dataset, we were able to identify five different classes of the substrate composition, including sponges, mixed submerged aquatic vegetation, mixed detritic bottom (fine and coarse) and unconsolidated bare sediment. We computed estimates of accuracy (namely Overall, User and Producer Accuracies) by cross tabulating predicted and reference instances. Overall, pixel based segmentations produced the highest accuracies and that the accuracy assessment is strongly dependent on the choice of classes for the segmentation. Tidal channels in the Venice Lagoon are extremely important in terms of habitats and sediment distribution, particularly within the context of the new tidal barrier being built. However, they had remained largely unexplored until now, because of the surveying challenges. The application of this remote sensing approach, combined with targeted sampling, opens a new perspective in the monitoring of benthic habitats in view of a knowledge-based management of natural resources in shallow coastal areas.
KW - Benthic habitat mapping
KW - high-resolution sonar
KW - image segmentation
KW - very shallow water
KW - multibeam
KW - Venice Lagoon
UR - http://dx.doi.org/10.1016/j.ecss.2015.12.014
U2 - 10.1016/j.ecss.2015.12.014
DO - 10.1016/j.ecss.2015.12.014
M3 - Article
VL - 170
SP - 45
EP - 60
JO - Estuarine, Coastal and Shelf Science
JF - Estuarine, Coastal and Shelf Science
SN - 0272-7714
ER -