Using context to detect underwater objects

In the absence of sufficient information about theidentity of an object to detect it among other objects thereis often extra information in the relationship between objectsthat can be exploited. This contextual information can helpus to characterise the natural set of surrounding objects oreven separate objects into different classes directly. Context iswidely used to understand and characterise scenery. This paperintroduces the use of context to detect underwater objects.Context has not previously been studied for complex sonarimagery and our aim is to design and test measures of contextto demonstrate how it can help us to detect and/or identify“unauthorised” objects that are not part of the natural set ofobjects present in the imagery. This is motivated by some ofthe challenges that are faced in cluttered sonar environments,or when the objects to be detected are hidden or disguised, forwhich extra information can be decisive.