Coastline extraction plays important functions in coastal resource management, natural resource preservation, and sustainable coastal development. Long-term records of Landsat data series are available for free downloading, being highly potential for coastline extraction. Furthermore, remote sensing imagery systems along with the application of digital image processing techniques can be utilized in coastline extraction. However, it is challenging to accurately extract coastlines with coarse spatial resolution due to the dynamic properties of various types of coastlines produced by sea-level changes from tides and storms. Moreover, the use of conventional surveys and the need for high-resolution images involve substantial costs and efforts, especially when coastlines of long distances are involved. The current study compared the accuracy of extracting coastlines from Landsat 8 OLI with the observed coastlines from GPS data. In particular, this study focused on distinguishing water areas from non-water areas on the coastline of a selected concrete port. The analysis involved the use of both MNDWI and NDWI indexes. The statistical results demonstrated the accuracy of MNDWI (RMSE of 2.363) and NDWI (RMSE of 2.491 m), which suggested the potential of using Landsat 8 OLI in coastline identification.