Research On Degraded Image Enhancement Algorithm Based On Dark Channel Prior

The display quality of images collected outdoors is severely degraded due to the interference of uneven light,fog,haze,sand and dust.Such images are collectively referred to as degraded images.Degraded images generally have the shortcomings of reduced image clarity and contrast,color attenuation,and noise amplification,which cause great obstacles to subsequent image processing,such as image recognition and feature extraction.Therefore,it is of great practical significance to develop an efficient degraded image enhancement algorithm to improve the display quality of the image.This thesis aims at enhancing the display quality of degraded images in foggy and sandy dust environments,and improves the traditional dark channel prior defogging algorithm to obtain an improved dark channel prior defogging algorithm,and compare it with the limited contrast adaptive histogram equalization algorithm,gray world assumption algorithm and gamma correction are weighted and fused,and a single image fusion defogging algorithm based on dark channel prior and a sand and dust image enhancement algorithm based on dark channel prior are respectively proposed.The algorithm in this paper is compared with other existing algorithms in simulation experiments,combined with visual effects and objective parameters to evaluate the effectiveness of the algorithm.The specific research content is as follows:1.Improved dark channel prior defogging algorithmBased on the traditional dark channel priori defogging algorithm,the local coordinate atmospheric light method is proposed.This method can improve the estimation accuracy of the atmospheric light value.At the same time,it uses the tolerance method to re-estimate the transmission of the bright area of the foggy image.It avoids halo,artifacts and color distortion in the image after defogging.The improved dark channel priori defogging algorithm effectively makes up for the shortcomings of the traditional dark channel priori defogging algorithm,laying the foundation for the later image defogging and enhancement.2.Single image fusion defogging algorithm based on dark channel priorThe algorithm successively uses the limited contrast adaptive histogram equalization algorithm and the improved dark channel prior defogging algorithm to obtain the contrast-adjusted image and the defogging image respectively,and then use the weighted fusion algorithm to fuse the above two images.By judging whether the variance is no longer increasing to obtain the best fusion coefficient,and get a higher quality fusion image,finally use gamma correction to increase the exposure of the image,restore the details of the image in the low-brightness area,and reduce the color difference between the images.The effectiveness of this algorithm is verified through comparative simulation experiments.In terms of visual effects,this algorithm can effectively improve the display quality of the image,and the overall visual effect of the image after defogging is better.In terms of objective parameters,this algorithm is significantly better than other algorithms,further verifying that the algorithm has a good dehazing effect.3.Sand and dust image enhancement algorithm based on dark channel priorCompared with foggy images,sand and dust images have serious chromatic aberration problems.Therefore,this algorithm first adjusts the chromatic aberration of the image using the gray world assumption algorithm in the Lab color space,and then uses gamma correction to adjust the dynamic range of the image to avoid noise amplification and color distortion.The improved dark channel priori algorithm defogs the corrected image,and then uses the limited contrast adaptive histogram equalization algorithm and the brightness compensation algorithm to improve the clarity and brightness of the defogged image,and finally weighted fusion of the above two types of images using a weighted fusion algorithm to obtain the final sand and dust enhanced image.The effectiveness of the algorithm is verified through simulation and comparison experiments.In terms of visual effects,the algorithm can effectively improve the color difference of the image,the clarity and visibility are significantly improved,and the overall visual effect is better.In terms of objective parameters,this algorithm is significantly better than other algorithms,further verifying that this algorithm can effectively improve the display quality of sand and dust images.