Single Image Dehazing Algorithm Based On Improved Dark Channel Prior And Guided Filter

Currently, with the development of computer vision and image processing technology,image dehazing has been an important researching direction in computer vision. Images acquired in bad weather, such as fog and haze are seriously degraded by the scatting of the atmosphere, which makes the color and contrast of the image degraded and makes the system relayed on optical imaging not worked. How to achieve better sharpness and contrast of the recovered image from a single hazy image is an important issue to be resolved. This paper focuses on single image dehazing algorithm based on improved dark channel prior and guided filter, the specific contents are as follows:Firstly, aim at the problem that some hazes cannot be removed using guided filter, we propose a single image dehazing algorithm based on roiling guided filter and improved dark channel prior. Atmosphere veil is obtained after guided filtering on the mixed dark channel which is defined based on dark channel. The initial transmission map can be obtained after getting the atmospheric light with the optimized atmosphere veil. Then roiling guided filter is addressed to solve the poor smoothness of the initial transmission map. The experimental results show that the proposed method has a very fast implementation and can resolve the problem that some hazes cannot be removed of the dehazed images.Secondly, to tackle the problem that dark channel prior is invalid for large sky or bright objects regions of single hazy image. A single image dehazing algorithm based on contextual regularization method is proposed. The coarser estimate of atmosphere veil is obtained after a map processing on the mixed dark channel. Then the contextual regularization method is utilized to solve the poor smoothness and edge-preserving of the initial transmission map. The experimental results show that the dehazed image with the proposed method has better sharpness, even for large sky or white objects regions.Finally, in order to get higher quality of the recovered image, a single image dehazing algorithm based on total variation regularization is proposed. Based on the initial transmission map we have obtained in the previous chapters. The total variation method isaddressed to solve the problem of poor smoothness of the initial transmission map. The experimental results show that the recovered image with the proposed method has better sharpness and visibility.