Ocean surveyors have historically relied on subsea video cameras deployed in underwater vehicles for environmental monitoring, general inspection, and underwater situational awareness. But the operating mechanism of “video” significantly limits the ability of advanced image processing and machine vision. Conversely, stills imaging provides revolutionary advantages when used correctly.
Voyis endeavours to produce underwater imagery where the effects of the water medium are entirely removed. Which will enable the use of standard terrestrial processing algorithms like photogrammetry and machine learning. In fact, this is a core focus of our mission to See the Depths like We See the Surface.
This blog is part of our Colour Imaging series. This edition will focus on our dynamic range and even illumination technology and how it can improve results.
A core challenge with underwater imaging is producing an even illumination across the entire camera field of view. This is critical both for the visual appearance of the image, but also to ensure that features can be detected both in the center of the image and the corners. Photogrammetry must be able to detect features across the entire imaging area to be effective.
This problem is driven by the fact that water absorbs light, and the degree of absorption is range dependent. Since the light illumination directed towards the corners of the image must travel further through the water, more of it is absorbed. So, even if light is distributed evenly across the camera the corners of the image will still be darker.
The diagram below illustrates this geometry. The light travelling to the image corners (corner light) must travel a longer total distance than the light illuminating the center of the image (center light).
Even Illumination & Dynamic Range
Voyis imaging systems employ a two-pronged approach to combat the phenomenon of uneven lighting, utilizing both physical and algorithmic lighting correction. The Nova Recon has each LED oriented at a specific angle, and each side is tilted outward in the perpendicular direction. This biases light to the front and edges of the image where absorption is most significant. This hydrodynamic external lightbar can be installed on the exterior of an AUV hull at a defined distance from the camera module.
But it has been found that even this light distribution biasing is not enough to provide perfectly even illumination, so the second prong is utilizing dynamic range. Dynamic range is the ability of the camera to resolve a wide range of light intensities, both very dim light returns and bright reflections.
A typical camera image is captured and displayed with 8-bits levels of brightness for each pixel – 255 levels of intensity. Voyis utilizes 12-bit or 16-bit imaging sensors, meaning that each pixel has 4,096 or 65,000 levels of intensity, respectively.
The image comparison below illustrates this capability. It demonstrates that the dark corners of the raw image can be enhanced without significant quality loss in its 12-bit form. On the other hand, an 8-bit enhancement is unable to resolve any features.
An image mosaic combines these light and distortion corrected images into one. The consistent lighting between sequential images makes it impossible to identify where the images have been stitched together.