Model-based super-resolution reconstruction techniques for underwater imaging

Abstract

The visibility of underwater imaging has been of long-standing interest to investigators working in many civilian and military areas such as oceanographic environments, efforts such as image restoration techniques can help to enhance the image quality; however, the resolution is still limited. Image super resolution reconstruction (SRR) techniques are promising approaches for improving resolution beyond the limit of hardware; furthermore, with the prior knowledge of the imaging system such as the point spread function and diffration limit, performance of the super resolution reconstruction can be further enhanced, which can also extend the imaging range as well. In order to improve the resolution to a best possible level, an imaging model based on beam propagation is established and applied to image super-resolution reconstruction techniques for an underwater range-gated pulsed laser imaging system in the presented effort. Experimental results show that the proposed approaches can effectively enhance the resolution and quality of underwater imaging