Change detection on shipwrecks using synthetic aperture sonar – North Sea Wrecks Task 3.5 Deep Water Case Study

During the last 15 years, synthetic aperture sonar (SAS) technology has matured substantially. Today SAS represents state-of-the-art in seabed imaging and mapping when used on autonomous underwater vehicles (AUV). SAS provides high resolution and large area coverage rate, and is well suited for a number of different applications in marine research, offshore survey mapping and monitoring, search for objects, wrecks and dumpsites, and military applications such as naval mine countermeasures and intelligence, surveillance and reconnaissance, and seabed warfare. The Norwegian Defence Research Establishment (FFI) has a long-standing collaboration with the Norwegian company Kongsberg Maritime (KM) to develop AUV-technology and SAS-technology. Today, there are multiple products in the HUGIN family of AUVs and the HISAS family of SAS systems available from KM, partly developed at FFI. Change detection (CD) is a technique to find relevant changes in data products collected in repeated passes. SAS, providing high resolution and large area coverage rate, is well suited for image based CD. In this feasibility study, we consider the suitability of using AUV with SAS for detecting structural changes on shipwrecks, with years of time separation. The potential application is monitoring of shipwrecks and their conditions in dumpsites. We describe the specific challenges related to both SAS imaging, mapping and CD of shipwrecks. We consider three data sets from 2015, 2019 and 2022 collected by FFIs HUGIN AUV carrying a HISAS 1032 interferometric SAS of wreck 13 in the Skagerrak World War II chemical munitions dumpsite. Summarized, using our approach, we are able to successfully coregistrate local images, produce difference images and detect small changes. To verify that the detected changes are real changes, the similarity in the imaging geometry for the repeated passes must be investigated. Shadow regions should be flagged and ignored in the change maps. Our conclusion is that AUV with SAS can be used for long term monitoring of shipwrecks if requirements are met regarding the similarity in the data collection and the sensor, and the choice of processing. This study is sponsored by the EU Interreg project North Sea Wrecks.