A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Multispectral satellite imagery and machine learning for the extraction of shoreline indicators
https://orcid.org/0000-0001-6479-0054
Abstract Analysis of shoreline change is fundamental to a broad range of investigations undertaken by coastal scientists, coastal engineers, and coastal managers. Multispectral Satellite Imagery (MSI) provides high resolution datasets that allow coastlines to be monitored more frequently and on a global scale. The Landsat and Sentinel-2 MSI datasets are free for public use, which has increased the frequency of studies focusing on coastal change using satellite imagery. However, despite access to global and free satellite imagery, a method has yet to be developed to monitor different shoreline types and indicators globally, as not all shorelines are sandy beaches, and the waterline cannot be representative of all shoreline changes. The review paper introduces different techniques used currently to extract shoreline features, including water indexing, Machine Learning (ML) and segmentation methods. We presented here a comprehensive review of range of the methods available for shoreline extraction from MSI and discuss why some shoreline features have been identified using multispectral satellite imagery and others not. This approach helps to signal where the gaps are on the current methods for shoreline extraction and provides a roadmap of the key challenges that prevents MSI to be used for understanding shoreline changes at a global scale.
Highlights Assessing shoreline indicators and their extraction through the use of Multispectral Satellite Imagery (MSI). There is a no “one size fits all” technique to monitor different shoreline types and indicators globally. Machine Learning can reduce the number of options for shoreline extraction for all temporal and spatial scales.
Multispectral satellite imagery and machine learning for the extraction of shoreline indicators
https://orcid.org/0000-0001-6479-0054
Abstract Analysis of shoreline change is fundamental to a broad range of investigations undertaken by coastal scientists, coastal engineers, and coastal managers. Multispectral Satellite Imagery (MSI) provides high resolution datasets that allow coastlines to be monitored more frequently and on a global scale. The Landsat and Sentinel-2 MSI datasets are free for public use, which has increased the frequency of studies focusing on coastal change using satellite imagery. However, despite access to global and free satellite imagery, a method has yet to be developed to monitor different shoreline types and indicators globally, as not all shorelines are sandy beaches, and the waterline cannot be representative of all shoreline changes. The review paper introduces different techniques used currently to extract shoreline features, including water indexing, Machine Learning (ML) and segmentation methods. We presented here a comprehensive review of range of the methods available for shoreline extraction from MSI and discuss why some shoreline features have been identified using multispectral satellite imagery and others not. This approach helps to signal where the gaps are on the current methods for shoreline extraction and provides a roadmap of the key challenges that prevents MSI to be used for understanding shoreline changes at a global scale.
Highlights Assessing shoreline indicators and their extraction through the use of Multispectral Satellite Imagery (MSI). There is a no “one size fits all” technique to monitor different shoreline types and indicators globally. Machine Learning can reduce the number of options for shoreline extraction for all temporal and spatial scales.
Multispectral satellite imagery and machine learning for the extraction of shoreline indicators
https://orcid.org/0000-0001-6479-0054
Abstract Analysis of shoreline change is fundamental to a broad range of investigations undertaken by coastal scientists, coastal engineers, and coastal managers. Multispectral Satellite Imagery (MSI) provides high resolution datasets that allow coastlines to be monitored more frequently and on a global scale. The Landsat and Sentinel-2 MSI datasets are free for public use, which has increased the frequency of studies focusing on coastal change using satellite imagery. However, despite access to global and free satellite imagery, a method has yet to be developed to monitor different shoreline types and indicators globally, as not all shorelines are sandy beaches, and the waterline cannot be representative of all shoreline changes. The review paper introduces different techniques used currently to extract shoreline features, including water indexing, Machine Learning (ML) and segmentation methods. We presented here a comprehensive review of range of the methods available for shoreline extraction from MSI and discuss why some shoreline features have been identified using multispectral satellite imagery and others not. This approach helps to signal where the gaps are on the current methods for shoreline extraction and provides a roadmap of the key challenges that prevents MSI to be used for understanding shoreline changes at a global scale.
Highlights Assessing shoreline indicators and their extraction through the use of Multispectral Satellite Imagery (MSI). There is a no “one size fits all” technique to monitor different shoreline types and indicators globally. Machine Learning can reduce the number of options for shoreline extraction for all temporal and spatial scales.
Multispectral satellite imagery and machine learning for the extraction of shoreline indicators
McAllister, Emma (author) / Payo, Andres (author) / Novellino, Alessandro (author) / Dolphin, Tony (author) / Medina-Lopez, Encarni (author)
Coastal Engineering ; 174
2022-02-27
Article (Journal)
Electronic Resource
English
3-D Shoreline Extraction from IKONOS Satellite Imagery
| Online Contents | 2003
Automatic Ship Detection in Satellite Multispectral Imagery
| Online Contents | 1993