Performance Evaluation of Shoreline Extraction Methods Based on Remote Sensing Data
Journal of Geography, Environment and Earth Science International • 2017
Publication Information
Authors
Sameh B. El Kafrawy; Mohamed E. Basiouny;Essam A.Ghanem; Ahmed S. Taha, 3*
Keywords
Shoreline; shoreline detection; coastal zones monitoring; Ras El-Hekma.
Journal
Journal of Geography, Environment and Earth Science International
Publisher
SCIENCEDOMAIN international www.sciencedomain.org
Volume
11
Issue
4
Pages
1-18
publication.type
International
Paper Link
Open Link
Supplementary Materials
Not Available
Abstract
Coastal zones monitoring is an important task in sustainable development and environmental protection. For coastal zones monitoring, shoreline extraction is a fundamental work. The shoreline
is defined as a line of contact between land and water body. The main purposes of this study were to compare and evaluate six of shoreline extraction methods based on remotely sensed data.
Those methods were digitizing, Thresholding band ratio, and classification, Normalized difference water index (NDWI), Normalized Difference Vegetation Index (NDVI), and Improved method. Those
methods were applied on landsat8 2015 image to extract the shorelines position for Ras El-Hekma coastal zone. The extracted shorelines were compared with reference 2015 shoreline which
detected by high resolution image Pleiades B1 (0.50 m) special resolution. A comparative study has been done depending on four evaluation criteria. The results show that all shorelines which are
extracted by six methods are accepted which 90% of shifting distances with in one pixel (30 m), but The shoreline extracted by Thresholding band ratio method is the most one closed to the true
shoreline where the mean value is 11.04 m, RMSE is 9.54 m about one third the value of the pixel size (30 m) of landsat 8 image, and 94.44% of the shoreline segments lie within 1 pixel distance of the true shoreline. The fourth evaluation criteria D90% is 22.35 m, which 90% of shifting distances between extracted and true shoreline is ling in.
is defined as a line of contact between land and water body. The main purposes of this study were to compare and evaluate six of shoreline extraction methods based on remotely sensed data.
Those methods were digitizing, Thresholding band ratio, and classification, Normalized difference water index (NDWI), Normalized Difference Vegetation Index (NDVI), and Improved method. Those
methods were applied on landsat8 2015 image to extract the shorelines position for Ras El-Hekma coastal zone. The extracted shorelines were compared with reference 2015 shoreline which
detected by high resolution image Pleiades B1 (0.50 m) special resolution. A comparative study has been done depending on four evaluation criteria. The results show that all shorelines which are
extracted by six methods are accepted which 90% of shifting distances with in one pixel (30 m), but The shoreline extracted by Thresholding band ratio method is the most one closed to the true
shoreline where the mean value is 11.04 m, RMSE is 9.54 m about one third the value of the pixel size (30 m) of landsat 8 image, and 94.44% of the shoreline segments lie within 1 pixel distance of the true shoreline. The fourth evaluation criteria D90% is 22.35 m, which 90% of shifting distances between extracted and true shoreline is ling in.
Staff Members - Benha University