Multiple polynomial regression for solving atmospheric scattering model. (2019)

• Record Type: Journal Article
• Authors: Verma, Monika; Yadav, Vikash; Kaushik, Vandana Dixit; Pathak, Vinay Kumar
• Is Part Of: International journal of advanced intelligence paradigms. Volume 12:Number 3/4(2019)
• Page Start: 400
• Page End: 410
• Publication Date: 2019
• Subjects: atmospheric scattering model -- ASM -- regression -- peak signal to noise ratio -- entropy; Artificial intelligence -- Periodicals; Machine theory -- Periodicals; Fuzzy logic -- Periodicals; 006.305
• Languages: English
• ISSNs: 1755-0386

Abstract

Multiple polynomial regression (MPR) is a novel technique for removing haze from images. Haze removal is a challenging task because not much information is there if the single image haze removal is taken into consideration. MPR is a method that solves the atmospheric scattering model (ASM). In solving the ASM from a single image many unknown variables like depth of the scene, scattering coefficient, transmission of the medium, atmospheric light have to be calculated. If all these variables are calculated with precision then it is possible to obtain a haze free image. The current work has shown improvements over the state of the art methods.