AN OBJECT BASED IMAGE ANALYSIS APPROACH FOR THE EXTRACTION OF THE KOLOUMBO VOLCANO AND ASSOCIATED DOMES-CONES FROM A DIGITAL SEABED ELEVATION MODEL

Bibliographic Details
Authors and Corporations: Βampourda, D., Argialas, D., Nomikou, P., Tzotsos, A.
Title: AN OBJECT BASED IMAGE ANALYSIS APPROACH FOR THE EXTRACTION OF THE KOLOUMBO VOLCANO AND ASSOCIATED DOMES-CONES FROM A DIGITAL SEABED ELEVATION MODEL
In: Bulletin of the Geological Society of Greece, 50, 2017, 3, p. 1616
published:
National Documentation Centre
Physical Description:1616
ISSN/ISBN: 2529-1718
0438-9557
Summary:<jats:p>This paper concerns the study of the seafloor through digital seabed elevation models, using object based image analysis methods. The goal of this research was the automated extraction of geomorphological features from the seabed in regions presenting intense volcanic activity. The study area is located around the submarine volcano of the Kolοumbo (in the submarine area northeast of the Santorini island, Greece). For this purpose, a Digital Elevation Model (DEM) of the seabed with a spatial resolution of 50m was used. Derivatives of the DEM, such us Slope, Topographic Position Index (TPI) and Terrain Ruggedness Index (TRI) were created in the open source software "QGIS 2.4". The implementation of the object based image analysis approach was performed in eCognition 8.7 software. Nine segmentation and classification levels were created in order to produce the final level segmentation "level 5", where the final geomorphological features were classified. The results of the method were evaluated using classification stability measures and qualitative and quantitative comparison of the results with existing map.</jats:p>
Type of Resource:E-Article
Source:National Documentation Centre (CrossRef)
Language: Undetermined