"Removing the water": physically-accurate color correction algorithm for underwater photography

Researcher Derya Akkaynak developed a photogrammetry algorithm that corrects underwater photography to remove color haze created by backscattering -- in effect "removing the water" to show the underwater world in its fullness of color and detail. The results are beautiful and crystal clear.

Why do all the pictures you take underwater look blandly blue-green? The answer has to do with how light travels through water. Derya Akkaynak, an oceangoing engineer, has figured out a way to recover the colorful brilliance of the deep.

You can read the scientific paper at OpenAccess: Sea-thru: A Method For Removing Water From Underwater Images [PDF]

The Sea-thru method estimates backscatter using the dark pixels and their known range information. Then, it uses an estimate of the spatially varying illuminant to obtain the range-dependent attenuation coefficient. Using more than 1,100 images from two optically different water bodies, which we make available, we show that our method with the revised model outperforms those using the atmospheric model. Consistent removal of water will open up large underwater datasets to powerful computer vision and machine learning algorithms, creating exciting opportunities for the future of underwater exploration and conservation.

A funny problem with the video: at 2:56 in, they're seen using Photoshop's color balance tools to edit one of the images, while Akkaynak says "this method is not photoshopping an image." It's surely just a terrible edit and they were doing a comparison, but it made me wonder, so I fired up Photoshop to see how it does.

Here, the paper's example is seen raw (top left), corrected by Photoshop's Auto Color tool (top right), corrected by Sea-Thru (bottom left) and corrected by me using the color balance sliders in Photoshop (bottom right). Read the rest