Conclusions and Future Directions

Abstract

This book leads us to efficient algorithms for fog from images and videos. The book started with the motivation for the development of fog removal algorithm. In Chapter 2, the atmospheric condition referred as fog is analyzed. Chapter 3 has described the sources where the foggy images and videos are available and the metrics used to measure the efficacy of the fog removal techniques. In Chapter 4, the development of fog removal algorithms for images and videos are described. In Chapter 5, a novel and efficient fog removal algorithm is presented for images captured by a single, uncalibrated camera system. Fog formation is due to attenuation and airlight. Attenuation reduces the contrast and airlight increases the whiteness in the scene. The presented algorithm uses anisotropic diffusion to recover scene contrast. Simulation results demonstrate that a presented anisotropic diffusion-based algorithm outperforms prior state of the art algorithms in terms of contrast gain, percentage of the number of saturated pixels and computation time. The presented algorithm is independent of the density of the fog and does not require user intervention. It can handle color as well as gray images. Along with the RGB color space, the presented algorithm can work on HSI color space which further reduces the computation.