Shoreline extraction from the fusion of LiDAR DEM data and aerial images using mutual information and genetic algrithms

As sea level rises and coastal populations continue to grow, there is an increased demand for understanding the accurate position of the shorelines. The automatic extraction of shorelines utilizing the digital elevation models (DEMs) obtained from light detection and ranging (LiDAR), aerial images and multi-spectral images has become very promising. In this paper, we propose a new algorithm that can effectively extract shorelines from fused LiDAR DEMs with aerial images depending on the availability of training data. The LiDAR data and the aerial image are fused together by maximizing the mutual information using the genetic algorithm. The extraction of shoreline is obtained by segmenting the fused data into water and land by means of the support vector machines classifier. Compared with other relevant techniques in literature, the proposed method offers better accuracy in shoreline extraction.