A correlation-based autofocus algorithm for coherent circular synthetic aperture sonar

摘要

Circular synthetic aperture sonar (CSAS) is a technique for generating imagery from scenes that have been ensonified by a sonar platform moving in a circular trajectory. While CSAS has many benefits over linear SAS from a resolution and target recognition standpoint, the generation of coherent, focused CSAS imagery without the use of a position reference in the form of a beacon has been considered a difficult problem due to motion and bathymetric constraints. An autofocus algorithm has been developed that utilizes many correlations between adjacent, narrow sub-aperture images to estimate a phase correction which can be used to form a focused, coherent CSAS image.