Shoreline and Coastline Extraction Using Multispectral UAS Imagery—A Case Study.

Shoreline and coastline are two important features of the coastal landscape that reflect the natural processes of erosion as well as deposition. Hence, extracting this information has become decisive for the study and analysis of the coastal region including its management and monitoring. Several methods of choosing and extracting the coastline and shoreline were analyzed from various research papers. The manual process of extraction is time-consuming and not efficient. Since automating the extraction using satellite images is ideal but expensive, the purpose of this research is to propose an effective low-cost methodology to achieve easy, accurate, and quick extraction of the coastal features. The study area considered for this research is the 120-acre Jupiter Inlet Lighthouse Outstanding Natural Area located in Jupiter, Florida, the USA that has more than 1,000 m (3,500 ft) of dynamic shoreline and coastline individually. Over the past few decades, significant erosion has affected this area due to various natural and anthropogenic factors. In this research, the extraction process was improved by using unmanned aerial systems based multispectral images, especially through RedEdge Micasense sensor to perform Modified Normalized Difference Water Index approach. Shoreline features have been generated through segmentation; vectorization followed by extraction processes in a short span of time without much human intervention. The extracted shoreline was compared with respect to the manually digitized shoreline using transect-based analysis to measure its accuracy [ABSTRACT FROM AUTHOR]