Side looking sonar image quality assessment using reference targets
conference paper
Side-looking sonar systems, including side-scan sonar and synthetic aperture sonar, are widely used for seabed characterization. They also have applications in the detection and classification of objects on the seabed. In mine hunting, contacts obtained after detection and classification have a corresponding sonar contact confidence level based on shape, size, shadow, structure, and target strength. These five criteria are linked to three metrics that characterise the quality of side-scan sonar imagery: geometric correctness, geometric resolution, and contrast. This paper focusses on the geometrical resolution, since it is one of the most important aspects of a sonar system to successfully detect and classify objects. To decide on a suitable sonar type and model, sonar specifications are an important source of information. Side-looking sonars are frequently mounted on autonomous systems and these systems can be operated with different settings (sonar mode, flying height, speed). Sonar image quality depends on the system characteristics (platform and sensor) and is influenced by the operational environment and method of deployment. In this paper, the development of a procedure to derive information on geometrical resolution using test targets is described. A point-spread-function is fitted through the backscattered response of the spherical targets. From which the resolution is derived as a function of range and flying height. The methodology is illustrated with side-scan sonar data collected with two autonomous underwater vehicles. The results show that the use of the developed test target and processing method allows to find the in-situ resolution.
TNO Identifier
858066
ISSN
14786095
ISBN
9781510872646
Publisher
Institute of Acoustics
Source title
Proceedings of the Institute of Acoustics, 4th International Conference on Synthetic Aperture Sonar and Synthetic Aperture Radar 2018, 5-7 September 2018, Lerici, Italy
Collation
8 p.
Pages
191-198
Files
To receive the publication files, please send an e-mail request to TNO Repository.