“Fast efficient algorithm for enhancement of low lighting video” by Dong, Pang and Wen

We describe a novel and effective video enhancement algorithm for low lighting video. The algorithm works by first inverting the input low-lighting video and then applying an image de-haze algorithm on the inverted input. To facilitate faster computation and improve temporal consistency, correlations between temporally neighboring frames are utilized. Simulations using naive implementations of the algorithm show good enhancement results and 2x speed-up as compared with frame-wise enhancement algorithms, with further improvements in both quality and speed possible.

Hau Ngo, Li Tao, Ming Zhang, Adam Livingston and Vijayan Asari. A Visibility Improvement System for Low Vision Drivers by Nonlinear Enhancement of Fused Visible and Infrared Video. Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2005, pp. 25 Google ScholarDigital Library
Night Driver, Making Driving Safer at Night, Raytheon Company, available at: http://www.nightdriversystems.com/nightdriver.htmlGoogle Scholar