International Symposium on Applied Geoinformatics 2019, İstanbul, Turkey, 7 - 09 November 2019, pp.1-7
Obtaining timely information about shorelines plays vital role for monitoring coastal areas. This study performed a comparative analysis of two object-based approaches for extraction of shorelines from Latvian Gulf of Riga (GoR). GoR was selected as study area since it has a large variety of coastal resources and natural biodiversity which are severely affected by climate change and human influences in recent years. The proposed method exploits the potential of near infrared (NIR) band to delineate the shorelines. Mean-shift and object-based fuzzy segmentation methods have been utilized for segmentation and extraction of shoreline. A dataset consisting of 9 digital orthophoto images was used for evaluation of the proposed method. These images were provided by Latvian Geospatial Information Agency which were acquired on May 21, 2013. The results obtained from each method were compared with photogrammetrically evaluated shorelines. Digital Shoreline Analysis System (DSAS) has been used to carry out accuracy assessment which is tool of commercial GIS software.