Algorithm of reconstructing underwater object images distorted by surface waving

The possibility of the use of data on the rough sea surface relief with the goal of mitigating refraction distortions in the images of the sea bottom or underwater objects observed through the air-water interface is analyzed. Algorithms for reconstructing a regular image from its random realization recorded simultaneously with a random realization of surface slopes are suggested and approved. It is shown that refraction distortions can be completely suppressed only if spatial variations in the object's visible radiance are recorded exactly, whereas in the case of a receiver characterized by a finite angular resolution, light focusing by wind waves can cause irreversible loss of information about object features clearly seen in the absence of waving. To reduce the negative effect of finite angular resolution on the quality of a reconstructed image, the recording system must ensure a certain reserve in resolution.