Underwater Image Enhancement using GAN

摘要

The process of enhancing the distorted underwater images to clear image is known as Underwater image enhance- ment. Distorted images are the raw underwater images that taken from the deep portion of ocean, river etc by using different cameras. In general underwater images are mainly used in underwater robotics, ocean pasture and environmental monitor- ing, ocean exploration etc. The underwater image enhancement process is done by using underwater image dataset which includes the distorted images (raw underwater images) and the corre- sponding enhanced underwater images. Currently used image enhancement methods cannot provide sufficient satisfaction to the underwater image enhancement. So proposed a new method by using Generative Adversarial Network (GAN), which tries to produce more images from the dataset.