A Real Time Multi-thread Underwater Images Enhancement System

Underwater images provide the most information for man-machine collaboration of underwater operations, however severe image quality degradation due to light absorption and scattering in water medium with suspended matter greatly affects efficiency to preceive the surrounding environment. In this paper, we develop a system for fast underwater enhancement to assist efficient underwater operations by giving more detailed and clear vision information. It uses auto multi-thread method to squeeze computing ability of hardware by creating a optimal number of processing threads considering both consistency and real time. And it is flexible to apply other algorithms or methods with a few changes as it is a method irrelavant framework and it can improve their efficiency.