Research On Image Defogging Method Based On Dark Channel Prior

At present,the requirements of performing the defog operation on the foggy images are to effectively improve the contrast of the foggy images,while recovering the color characteristics of the foggy images and the detailed information contained in the images.At present,the most concerned in the research field of image defogging method is the dark channel prior algorithm.But the algorithm has the following two fatal flaws while effectively recovering the fogged image.First,this method not only involves the reversible operation of large matrices,which results in high time and space complexity,but also causes edge effect of image.Second,the dark channel prior theory will invalidate for large light areas of the image,which will make the recovered image too saturated and severely distorted.This article mainly does the following work for these problems:(1)Mainly study the theory and implementation process of the dark channel prior defogging algorithm.Firstly,the image defogging model is established.Then the parameters are estimated.Finally the parameter inversion is performed to obtain the defogging image.(2)In view of the high time complexity and the edge effect of the image of He's dehazing algorithm,this paper proposes an improved algorithm.First,use the four-point weighting algorithm to accurately estimate the atmospheric light value.Then guided filtering is used to refine the rough transmittance obtained by using the dark channel prior primary.Finally,substitute the transmittance and atmospheric light value into the atmospheric scattering model to obtain the image after defogging.(3)An improved algorithm is proposed for the problem that the dark channel prior theory fails for large areas of light-colored areas.First,the sky area and the non-sky area are separated by a method that combines edge detection and optimized maximum inter-class variance.Then the non-sky area uses the previous improvement method,and the sky area uses the sequence decomposition enhancement algorithm.Finally,the non-sky area after defogging and the enhanced sky area are fused to obtain a clear fog-free image.(4)Based on the effectiveness of the improved algorithms proposed in this paper,simulation experiments are conducted to verify the proposed improved algorithms from subjective visual evaluation and objective quantitative indicators.