Automatic extraction of shorelines from Landsat TM and ETM+ multi-temporal images with subpixel precision

Abstract

13A high geometric precision method for automated shoreline detection from Landsat TM and ETM+ 14 imagery is presented. The methodology is based on the application of an algorithm that ensures accurate 15 image geometric registration, and a new algorithm for sub-pixel shoreline extraction, both at sub-pixel 16 level. The analysis of the initial errors shows the influence of the differences in reflectance of land cover 17 types over the shoreline detection, allowing us to create a model to substantially reduce these errors. 18Three correction models were defined attending to the type of gain used in the acquisition of the original 19Landsat images. Error assessment tests were applied on three straight coast segments artificially 20 stabilized, all of them located in microtidal coastal areas. A testing set of 45 images (28 TM, 10 ETM 21 high-gain and 7 ETM low-gain) was used. The mean error obtained in shoreline location ranges from 1.22 22 to 1.63 m, and the RMSE from 4.69 to 5.47 m. Since the errors follow a normal distribution, then the 23 maximum error at a given probability can be estimated. The results obtained show the possibility to apply 24 this methodology over large coastal sectors in order to determine and analyse the evolution trend of these 25 dynamic areas. 26 27

Variabilidad De La Línea De Costa a Partir De Imágenes De Satélite Y Su Relación Con La Textura Del Sedimento

“…En los últimos años se ha realizado un importante esfuerzo investigador tratando de determinar de forma eficiente la posición de la línea de costa partiendo de imágenes espaciales de resolución media (Foody et al 2005;Muslim et al 2006;Pardo-Pascual et al 2012;Almonacid-Caballer 2014;Sánchez-García et al 2015;García-Rubio et al 2015;Liu et al 2017), permitiendo obtener una enorme cantidad de información con una elevada frecuencia de actualización. El empleo de estos algoritmos permite registrar los cambios en la posición de la línea de costa con una alta frecuencia a lo largo de un año ).…”

“…Así se ha dispuesto de 16 líneas de costa que cubren la totalidad de la zona estudiada. Se ha trabajado con los puntos de la línea de costa obtenidos de la banda 6 de Landsat 8, al considerar que es la más estable y que menor error presenta, según ha sido comprobado en trabajos previos (Pardo-Pascual et al 2012;Almonacid-Caballer 2014).…”

“…La obtención de las líneas de costa mediante SHOREX (Pardo-Pascual et al 2012;Almonacid-Caballer 2014) presenta un error estimado de 5 metros (EMC), evaluado sobre tramos fijos de escollera. Esta incertidumbre inherente al método de deducción de la línea de costa introduce ruido en la definición de la orilla, de forma que las variabilidades aquí analizadas presentan un componente independiente a la propia variabilidad experimentada por la línea de costa.…”

Investigation of Morphological Change at the Cua Dai River Mouth Through Satellite Image Analysis

“…The level of accuracy of shoreline extraction would always be worse than the 30 m resolution of the original images. However, that could be applicable for cases which have remarkable changes of magnitude such as river delta or river mouth (Pardo-Pascual et al, 2012). Several techniques to rectify and extract shoreline position from Landsat images and associated uncertainties can be found in Feyisaa et al (2014); Maiti and Bhattacharya (2009);Pardo-Pascual et al (2012) and Ryu et al (2002).…”

“…However, that could be applicable for cases which have remarkable changes of magnitude such as river delta or river mouth (Pardo-Pascual et al, 2012). Several techniques to rectify and extract shoreline position from Landsat images and associated uncertainties can be found in Feyisaa et al (2014); Maiti and Bhattacharya (2009);Pardo-Pascual et al (2012) and Ryu et al (2002). While, the technique and relevant uncertainties when extract shoreline position from high-resolution aerial photographs can be found from Pradjoko and Tanaka (2010).…”