Real-time autonomous multi resolution visual surveys based on seafloor scene complexity

This paper describes a method to optimize the spatial resolution of image surveys based on the spatial scale of features on the seafloor that are not known prior to observation. The method makes use of the density of visual features as a measure of the complexity of a seafloor image. In order to achieve this, two approaches to assess scene complexity we investigated. The performance of the method was verified using seafloor imagery obtained in the Iheya North field in the Okinawa Trough. The results demonstrate that it is effective for a large range of feature sizes.