Image Dehazing using Guided Image Filter
| Author(s) | : | Prof. Raskar V.B, Mr.Tupe S.G |
| Institution | : | Assistant Professor, Department of E&Tc Engineering, JSPM’s Imperial College of Engineering & Research, Pune |
| Published In | : | Vol. 4, Issue 7 — July 2017 |
| Page No. | : | 43-49 |
| Domain | : | Engineering |
| Type | : | Research Paper |
| ISSN (Online) | : | 2348-4470 |
| ISSN (Print) | : | 2348-6406 |
—Imaging in bad weather is frequently tarnished by dispersion due to suspended particles in the tropospheresuch as haze, fog and mist. It decreases scene visibility. In this paper we have proposed a haze removal algorithm usingcolor attenuation prior model and Guided Image Filter. By creating a linear model for modelling the scene depth of thehazy image under this novel prior and learning the parameters of the model by using a supervised learning method, thedepth information can be well recovered. Once depth information is recovered, it is used to estimate transmission map.We can easily restore the radiance via the atmospheric scattering model and thus perfectly remove the haze part from theimage. We have implemented this algorithm by using Guided image filter and Weighted Guided Image Filter. The use ofWeighted Guided Image Filter improves the performance of algorithm. The algorithm achieves high quality dehazingwith high structural similarity and less mean square error.
Prof. Raskar V.B, Mr.Tupe S.G, “Image Dehazing using Guided Image Filter”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 4, Issue 7, pp. 43-49, July 2017.
