Algorithm makes polarized imaging more practical for imaging through water, which could benefit rescues and other underwater applications
WASHINGTON - Researchers have developed a new method that can automatically produce clear images through murky water. The new technology could be useful for searching for drowning victims, documenting submerged archaeological artifacts and monitoring underwater farms.
Imaging clearly underwater is extremely challenging because the water and the particles in it tend to scatter light. But, because scattered light is partially polarized, imaging using a camera that is sensitive to polarization can be used to suppress scattered light in underwater images.
Caption: The researchers used a traditional polarization imaging setup to image various objects submerged in the turbid water. Combining this simple setup with a new algorithm produced clear underwater images without prior knowledge or a background region.
Credit: Haofeng Hu, Tianjin University
"Our new method overcomes the limitations of traditional polarimetric underwater imaging, laying the groundwork for taking this method out of the lab and into the field," said research team leader Haofeng Hu from Tianjin University in China. "Unlike previous methods, there's no requirement for the image to include a background area to estimate the backscattered light."
In The Optical Society (OSA) journal Optics Express, the researchers demonstrate their method's ability to enhance image contrast while preserving image details without introducing considerable noise. The new method even works in dense turbid water, which is so cloudy it is almost impossible to see through.
"Our polarimetric imaging method can improve the image quality in various scattering media, not just turbid water," said Hu. "We think the principle we used might be extendable to imaging through other scattering media such as fog, haze and smoke."
Practical underwater imaging
Traditional approaches to underwater imaging use either prior knowledge of the imaging area or the background of an image to calculate and remove scattered light. These methods have limited utility in the field because they typically require manual processing, images do not always have visible backgrounds, and prior information is not always available.
Caption: The researchers demonstrated their technique by imaging a scene on a wood board submerged in liquid with various turbidities. The left images were taken with a traditional camera, and the right images show the clear images produced by the new method.
Credit: Haofeng Hu, Tianjin University
