Object Classification and Segmentation Based on Deep Learning Using Underwater Mapping Data

Abstract

. This paper presents a fast and accurate classification method for underwater objects using underwater mapping data obtained by a small Autonomous Underwater Vehicle (AUV) and autonomous surface vehicle (ASV). For the mapping data, in addition to underwater acoustic reflection intensity images, water depth data, point cloud data and backscattering reflection intensity data are employed. We propose the automatic classification and semantic segmentation method on deep learning using a convolutional neural network (CNN) and PointNet++. In order to verify the effectiveness of the present method, we applied it to the measured several underwater mapping data.