Research On Defogging Algorithm Of Seaside Fog Image Based On Dark Channel Prior

The beach environment in fog,the water content in the suspended particles from the atmosphere is higher,the scattering of atmospheric light is stronger,the fog concentration is greater.The acquired image quality degrades,affecting the computer image analysis,identification and processing.Thus,dealing with the fog images effectively,increasing the sharpness of the images,improving the quality of the images,have the urgent need for reality and theory.In current,the research of image defog also have many methods,mostly based on the dark channel prior algorithm that make the effect of image defog remarkable.However,this method still exists some disadvantages like sky color distortion,image processing speed slow.In this paper,aiming at these disadvantages and combined with the characteristic of beach image,a kind of algorithm for beach image defog that based on dark channel prior is presented.Aiming at the phenomenon that the dark channel algorithm leads to color distorted when dealing with the sky area,this algorithm improves the estimation method of atmospheric light intensity.An improved K-means clustering algorithm is proposed to segment the original image for the beach image that contains more sky area,then receive the area of the sky whose image information is preserved relatively complete.The atmosphere light intensity of the whole image is obtained by non-sky area luminance values with the sky region luminance values weighted.Because soft matting algorithm uses long time to estimate transmittance,this paper adopts guide filter to optimize the transmittance.It can not only shorten the algorithm time,but also realize the edge smoothing,enhance details,effectively remove block-effect,get a relatively accurate transmission rate estimate.As the fog image of the beach is dark,the color is not obvious,image contrast is low,tone remapping is used to improve the image contrast for restored image.In this paper,the defog algorithm based on dark channel prior is simulated by MATLAB experiment platform,and compared with the median filter and the bilateral algorithm.The experimental results show that the algorithm can effectively remove the influence of fog on the beach image,make the image more natural and bright,the contrast is higher,the visibility is better,and the fog image can be processed quickly and effectively.