UC San Diego

Within the past decade, there has been a growing interest in the development of surveillance systems deployed in the maritime domain. Surveillance in the maritime domain is confronted with many problems that are not easily solved. Most problems are caused by the weather, such as fog, haze, and turbulence that degrade the contrast and image quality of images and video. Most recently, researchers have proposed methods to remove fog in images fast enough for real-time processing. However, they are not unified in their approach. Additionally, there exists no metric that indicates the perceptual quality of an image based on the contrast. In this work, we discuss the problems encountered when including contrast enhancements for fog removal along with image and video compression. We unify existing fog removal methods with our proposed Color Ellipsoid Framework and present a new fog removal method. We then utilize the Color Ellipsoid Framework to improve the performance of a no-reference perceptual based contrast enhancement metric. Over the years many researchers have provided insight into the physics of either the fog or turbulence but not both. In this work, we provide an analysis and method that incorporates both physics models : fog and turbulence. We observe how contrast enhancements (fog removal) can affect object tracking and frame averaging. We present in this work a new joint contrast enhancement and turbulence mitigation method (CETM) that utilizes estimations from the contrast enhancement algorithm to improve the turbulence removal algorithm. We provide two new turbulent mitigation metrics (based on the Color Ellipsoid Framework) that measures temporal consistency. And finally, we design the CETM to be efficient such that it can operate in fractions of a second for near real-time applications