Research Fog Clear Method Of Degraded Images

In recent years fog and haze appear frequently around the country. This will reduce the quality of the image and interfere the performance pictured by the computer system. The reliability of the monitoring, navigation, remote sensing which based on the outdoor system will be reducing. The pictured images occurs severe degradation, get image contrast lower, and color darker, impact the system performance to avoid the negative impact of the environment on the system, research on fog degraded image has a very important practical significance.This paper works on the degraded image which collected in the fog days, aimed at degraded image’s poor visibility, color distortion, make the image restoration and enhancement. Base on the atmospheric scattering model, explanation the fog image degradation. Focus on the principle to explain the experimental principle and the realization of the process of the image dehazing method based on dark channel prior.Because a number of problems in the defog method, we adopted the dark primary colors priori defogging algorithm based on the depth map of the scene. This algorithm used the scene depth changes in the relationship of the dark primaries repair dark map. While in the dark-channel prior district does not exist, add an adjustment factor to adjust the transmittance in processing, then the adaptive correction of the mobile module window, so that the algorithm can adapt to the different dimensions of the processed image. The experiments show that the algorithm greatly reduces the complexity of the algorithm, effectively reducing the time.After discovery through observation, in the calculation process it will be loss of image information. To solve this problem, we adopted a efficient fog removal. Proposed algorithm uses anisotropic diffusion for estimation of airlight. Proposed algorithm does not require user intervention. Results show that proposed algorithm performs well in comparison with other existing algorithms. Even in case of heavy fog, proposed algorithm performs well, as algorithm is independent of the density of fog present in image. By a large number of experiments and data, we verify the effectiveness of the proposed algorithm to fog.