A single picture defogging method based on fog layer smoothing priori

Abstract

The invention discloses a single image defogging method based on fog layer smoothing priori. According to the method, a new fog image composition model is provided according to the daily observation phenomenon, it is considered that formation of a fog image is a degraded image formed by covering a clear image with a fog layer, and therefore the purpose of the invention is to peel the clear image from the fog image. In order to recover a clear picture, fog layer smoothing priori is introduced for constraint. The priori believes that the fog layer covering the clear image is smooth, i.e., the gradient histogram of the fog layer is distributed in a short tail manner. Through analysis, the joint probability density function of the clear image gradient and the fog layer gradient is maximized, the smoothness of the fog layer can be well described, a corresponding model is provided for solving, and finally a defogged result is obtained. According to the method, a new fog graph composition structure is provided, the smooth prior of a fog layer is effectively utilized, the fog layer is separated, a clear image is obtained, and a defogging result with the peak signal-to-noise ratio and the structural self-similarity both higher than those of a current popular method is obtained.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a single image defogging method based on fog layer smoothness prior. Background technique [0002] As a natural phenomenon, fog has not only affected people's visual experience for a long time, but also is very unfavorable to outdoor activities such as traffic travel and aerial work. Images captured in foggy environments often suffer from low contrast, halos, and color shifts, which limit the visibility of the image. In recent years, with the continuous deterioration of the air environment, smog has become prevalent in major cities, which not only endangers the health of organisms, but also further reduces outdoor visibility, resulting in frequent traffic accidents and hindering outdoor activities. In addition, in the field of computer vision, many computer vision and image processing algorithms are not robust to foggy images, and the existence of fog hinders fu...

AI Technical Summary

Problems solved by technology

[0011] The difficulty to be overcome by the above various defogging methods is to accurately estimate the transmittance t and the global atmospheric light A;

[0012] Secondly, the recovery effect of the above-mentioned methods on de-fog images is far worse than that of light fog. fog formation

Images