PROJECT TITLE :

Image Haze Removal via Reference Retrieval and Scene Prior - 2018

ABSTRACT:

Photography of hazy scene usually suffers from low-contrast which degrades the visibility of the scene. The performance of single-image dehazing methods is limited by the priors or constraints. During this Project, we have a tendency to present an efficient methodology for haze removal, which utilizes its retrieved correlated haze-free images as external information. The correlated haze-free images are with scene previous providing scene structure and local high frequency info for dehazing, although variations in viewpoints, scales, and illumination conditions exist. To utilize those reference a lot of effectively, global geometric registration and native block matching toward the hazy input are performed to bolster the spatial correlations. Based on the registration, different types of external information are estimated. Furthermore, we combine that extra external info with internal constraint and regularization for estimating scene transmission map. Experiments demonstrate that our approach can produce dehazing results with higher visual quality compared with alternative state-of-the-art ways.