Airborne and Ground Experiment of Hyperspectral Remote Sensing and Applications to Ship Detection

Abstract

The hyperspectral sensor utilizes the reflection characteristics of objects by dividing it into several narrow and continuous hundreds of channels, unlike the multi-spectral sensor utilizing only the spectral information of the five or sixchannels. Because every pixel can be analyzed through hundreds of channels, the hyperspectral sensor is used in diverse oceanic phenomen. In this study, hyperspectral measurements were obtained using airborne and ground experiments and hyperspectral data processing techniques were applied for the detection of vessels. The observed hyperspectral data were analyzed by using the spectral characteristics of endmembers and a representative spectral mixture analysis technique. Thepixels in the image were classified into endmembers by using spectral matching algorithms and the marine library spectra from preliminary experiments. The estimated lengths and widths of the detected ships based on a series of hyperspectral data processing methods show RMS errors of 2.28% and 8.72% with respect to actual ship sizes, respectively. The investigation of the errors revealed that undulating surface waves, surface water penetration of the deck, and the submerged portion of the ship below the water surface potentially contribute to unclear ship boundaries. This study has significance in that the hyperspectral images is newly applied to the ship detection field.