Deep Learning-Based Maritime Environment Segmentation for Unmanned Surface Vehicles Using Superpixel Algorithms

Unmanned surface vehicles (USVs) are receiving increasing attention in recent years from both academia and industry. To make a high-level autonomy for USVs, the environment situational awareness is key capability. However, due to richness of features marine environments, as well complexity influenced by sun glare sea fog, development reliable system remains challenging problem that requires further studies. This paper, therefore, proposes new deep semantic segmentation model together with Simple Linear Iterative Clustering (SLIC) algorithm, an accurate perception various maritime environments. More specifically, powered SLIC can achieve refined results around obstacle edges improved accuracy water segmentation. The overall structure employs encoder–decoder layout, superpixel refinement embedded before final outputs. Three publicly available image datasets used this paper train validate model. output demonstrates proposed provide