Ship detection based on airborne and in situ hyperspectral measurements

Abstract

As maritime accidents around the Korean peninsula are increasing, the importance of ship detection research using remote sensing data is increasingly emphasized. This study presents a new ship detection algorithm using hyperspectral imagesthat provide spectral information of several hundred channels in ship detection field which depends on high resolution optical image. We applied a spectral matching algorithm between the reflection spectrum of the ship deck obtained from two field observations and the ship and seawater spectrum of the hyperspectral sensor of AVIRIS (Airborne Visible/Infrared Imaging Spectrometer). A total of five detection algorithms were used, including Spectral Distance Similarity (SDS), Spectral Correlation Similarity (SCS), Spectral Similarity Value (SSV), Spectral Angle Mapper (SAM) and Spectral Information Divergence (SID), and the proposed method shows the advantages and disadvantages of various detection algorithms such as SAM technique showing a clear classification, SDS technique in which a part of a ship is detected by seawater, and SID technique showing a noise distribution without a specific pattern in seawater. Also, we classified the ship of hyperspectral image by applying adaptive threshold value. As a result, SAM and SID algorithm showed superior ship detection ability than other methods.