Oceanographer develops algorithm to “remove” water from underwater photos

Derya Akkaynak is an oceanographer and engineer who focuses on imaging and underwater vision. Her newest project, Sea-thru, dives into how we can source clearer images of the underwater world. Akkaynak has developed an ingenious algorithm that essentially “removes” water from underwater photographs. 

The green-blue hue on underwater images is caused by what researchers call “backscatter.” This can be due to particles in the water, but it also occurs because the light is distorted and scattered as it travels through water. The resulting haze can make objects look distorted and obscures far away features. 

Akkaynak is traveling through the Lembeh Strait in Indonesia to test out the algorithm in action. When she comes across a large underwater feature, like a coral formation, she places a color chart at its base and photographs it from all angles using a regular underwater camera. Once she has captured it from all sides and measured the distances from the camera to the feature, she inputs this data and the corresponding images into her computer and applies her specific mathematical formula. 

The formula carefully analyzes each pixel and removes distortions in color and shape caused by the water. The resulting image is a clear picture of an underwater landscape. It looks almost as if you had drained the surrounding water. 

This new method has a wide range of practical applications. For marine biologists, it means clearer images of sea life for more precise analysis. It also allows scientists to more accurately assess changes in an underwater ecosystem. It’s even valuable for recreational divers looking to capture more specific images of the sea life they discover. 

Image source: Towards Data Science

Solution News Source

Oceanographer develops algorithm to “remove” water from underwater photos

Derya Akkaynak is an oceanographer and engineer who focuses on imaging and underwater vision. Her newest project, Sea-thru, dives into how we can source clearer images of the underwater world. Akkaynak has developed an ingenious algorithm that essentially “removes” water from underwater photographs. 

The green-blue hue on underwater images is caused by what researchers call “backscatter.” This can be due to particles in the water, but it also occurs because the light is distorted and scattered as it travels through water. The resulting haze can make objects look distorted and obscures far away features. 

Akkaynak is traveling through the Lembeh Strait in Indonesia to test out the algorithm in action. When she comes across a large underwater feature, like a coral formation, she places a color chart at its base and photographs it from all angles using a regular underwater camera. Once she has captured it from all sides and measured the distances from the camera to the feature, she inputs this data and the corresponding images into her computer and applies her specific mathematical formula. 

The formula carefully analyzes each pixel and removes distortions in color and shape caused by the water. The resulting image is a clear picture of an underwater landscape. It looks almost as if you had drained the surrounding water. 

This new method has a wide range of practical applications. For marine biologists, it means clearer images of sea life for more precise analysis. It also allows scientists to more accurately assess changes in an underwater ecosystem. It’s even valuable for recreational divers looking to capture more specific images of the sea life they discover. 

Image source: Towards Data Science

Solution News Source

SIGN UP

TO GET A Free DAILY DOSE OF OPTIMISM


We respect your privacy and take protecting it seriously. Privacy Policy