Investigating the evolution of Lebanese shoreline using object-based image classification - Université de La Réunion Access content directly
Conference Papers Year : 2018

Investigating the evolution of Lebanese shoreline using object-based image classification


The Lebanese coast has been experiencing increased anthropological pressures. The detection and monitoring of shoreline evolution are essentially needed in order to identify areas that require protection and/or further investigation and management intervention. Various methodologies have been developed to study changes in coastal areas. These included comparing historical aerial photographs and coastal topographic maps, 2) evaluating erosion, accretion, artificialization and seafilling of coastlines, 3) developing models to calculate sediment transport, and 4) identifying hazard prone coastal areas, amongst others. Remotely sensed data, including satellite imagery and aerial photography, has become an essential tool in numerous types of classification. Traditionally, most classifications have been created using a pixel-based analysis of remotely sensed imagery. With the growing availability of higher resolution imagery, this spatial information and its context could be used to produce more accurate classifications. This study aimed at detecting erosion, accretion, and seafilling along the Lebanese coast between 1962 and 2016. The specific objectives were to combine and investigate the use of multi-source imagery namely aerial photograph (0.50 m) Landsat 8 panchromatic (15 m) and Sentinel 2-A imagery (10 m) in an object-based image analysis approach, and 2) investigate the classification performance in detecting changes in the shoreline along the coastal zone. Geographic object-based image analysis (GEOBIA) was employed to map changes in the shoreline between 1962 (i.e., reference shoreline) and 2016 using aerial photography and Landsat 8 imagery, respectively. Similarly, changes were mapped between 1962 and 2016 using the same aerial photography and Sentinel 2-A imagery, respectively. Quickbird panchromatic imagery (0.6 m) was then employed in combination with the 1962 imagery for evaluation of the classification results. Accordingly, the shorelines in 1962 and 2010 were manually digitized using ESRI ArcGIS 9.2 software with the High Water Mark (HWM), defined as dry/wet line, adopted as the shoreline indicator. The evolution study was therefore conducted using simple photo-interpretation method. Surfaces were measured as polygon areas and then employed for evaluating the capability of the GEOBIA classification model in capturing changes along the coastal line for each Landsat 8 and Sentinel 2 A imagery. In this context, the results of detecting erosion, accretion, and sea-filling along the coast of Lebanon between 1962 and 2016 using both sources of imagery were discussed accordingly.
No file

Dates and versions

hal-01960304 , version 1 (19-12-2018)


  • HAL Id : hal-01960304 , version 1


George Mitri, Manal Nader, Manale Bou Dagher. Investigating the evolution of Lebanese shoreline using object-based image classification. GEOBIA 2018 - From pixels to ecosystems and global sustainability ​, Centre d'Etudes Spatiales de la BIOsphère (CESBIO); Office national d'études et de recherches aérospatiales (ONERA); Espace pour le développement (ESPACE DEV); Société T.E.T.I.S, Jun 2018, Montpellier, France. ⟨hal-01960304⟩
145 View
0 Download


Gmail Mastodon Facebook X LinkedIn More