Shoreline detection using remote sensing data with deep learning technology

Abstract

This paper presented method identifying shorelines using remote sensing images and deep learning method. Experiments were conducted at two areas in Vietnam using Sentinel-2 satellite imagery. Comparing deep learning with traditional methods, such as support vector machine (SVM) and NDWI index shows that deep learning is an effective method for semantic segmentation remote sensing data. Measures for improving the accuracy of deep learning model when determining water surface were also proposed.