An Improvement of Multi-Scale Fusion Technique for Image Dehazing
Megha Sara Alex1, Biju V. G2
1Megha Sara Alex, M.Tech Scholar, Department of Computer Science and Engineering, College of Engineering Munnar, Idukki, Kerala, India.
2Biju V. G, Assoc. Prof., Department of Electronic and Communication Engineering, College of Engineering Munnar, Idukki, Kerala, India.
Manuscript Received on May 08, 2015. | Revised Manuscript Received on May 09, 2015. | Manuscript published on May 20, 2015. | PP: 1-5 | Volume-3 Issue-6, May 2015. | Retrieval Number: F0626053615/2015©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The visibility of an outdoor imagery can be decayed by the hazy weather condition. In this paper, a modified multi-scale fusion technique is proposed to remove dense homogenous haze. The method first finds the airlight from the fused image. Then the patch-wise transmission function of the fused image is estimated by applying boundary constraints. After that a weighting function is applied on the patch-wise transmission to enhance the image structures. Using the patch-wise transmission and the weighting function, the transmission function is estimated. The dehazed image is hence recovered based on the correct estimation of both airlight and transmission function. The proposed method is then compared against the existing multi-scale fusion method by Qualitative and Quantitative evaluation. The results obtained show that the new method enhances the image visibility in dense hazy regions.
Keywords: Outdoor imagery, Dehazing, Multi-Scale fusion, Airlight, Transmission function.