EDGE AWARE WEIGHTED GUIDED IMAGE FILTERING FRAMEWORK FOR IMAGE ENHANCEMENT, DEHAZING AND IMAGE FUSION

Abstract

The general and conventional method of image filtering is filtering with edges but this method suffers from halo artifacts. In this proposed work, the use of aware weighting into an existing Image Guided Filter to sort out the problems such as halo artifacts. The WGIF has the advantage of both methods of filtering such as Global filtering and Local filtering. WGIF has many advantage as compared with conventional Guided Image Filter. Other advantages are, the WGIF removes halo artifacts in the same ways as that of the conventional Global filters and complexity of the WGIF is same as that of Guided Image Filtering. The WGIF can be applied for single image dehazing, single image detail enhancement and dehazing. This WGIF produce images with better visual quality, better color contrast and fidelity. Experimental results prove that the resultant algorithms can give images with better visual quality at the same time other artifacts can be removed. In future, this idea can be adapted for improvement of anisotropic diffusion and Poissons image editing[9].