Improved Dark Channel Prior Dehazing Algorithm Based On Atmospheric Scattering Model

Image dehazing is a very active research topic in the field of image processing,and also an important technology of modern image enhancement.It has a large amounts of applications,containing transportation,remote sensing,navigation,etc.Over recent years,various methods have been proposed for Image dehazing,Image dehazing has been rapid developed both in theoretical research and real-world applications.However,the key and difficult points of image dehazing still lie in the solution of atmospheric light value and transmission function.At the same time,it is necessary to make the dehazing image have good visual effect and protect more image information.Therefore,how to overcome these difficulties and obtain more accurate dehazing images is a great challenge for image dehazing researchers,which is also the focus of this study.In recent years,the dehazing algorithm based on dark channel priori has been widely used because of its superior fog removal effect.Firstly,this paper will introduce the dehazing algorithm based on dark channel prior,and on this basis,make corresponding improvements to each problem of the traditional dark channel prior dehazing algorithm.This paper mainly includes:1)The most important point of image defogging technology is the solution of transmission function.Firstly,tolerance mechanism will be used to select high brightness region in which the dark channel prior theorem fails.Then,in high brightness region where dark channel prior theorem fails,a method based on boundary constraints is used to obtain the rough transmission function,while in the non-high brightness region,the dark channel prior theorem is still used to solve the transmission function image.After that,considering the time complexity of matting algorithm in optimizing the transmission function,this paper uses guided filtering to optimize the rough transmission image.In high brightness region,there is a small problem in using the boundary constraint theory to obtain the transmission function.In the high brightness region,there is a problem that the value of the transmission function is small by using the boundary constraint theory.Therefore,this paper improves the formula of tolerance mechanism slightly,and then uses the formula of tolerance mechanism to enhance the transmission function of high brightness region.Finally,a more optimized image of transmission function is obtained.2)If we want to improve the visual recovery of dehazing image and work out the problem that the dehazing image is darkness,it is necessary to enhance the brightness of defogging image.For the high brightness area,a pseudo defogging image is defined,and the brightness of the high brightness area of the image is enhanced by fusing the defogging image with the pseudo defogging image.For the non-high brightness area in the image,a histogram equalization method is used to process the brightness component image of the dehazing image to enhance the brightness of the non-high brightness area.