Video image sharpness processing method in fog and haze day and device thereof

Abstract

The invention relates to a video image sharpness processing method in a fog and haze day and a device thereof. The method comprises the following steps: establishing a quantitative model between an atmospheric transmission parameter and an optical propagation medium parameter; obtaining the optical propagation medium parameter of a practical measurement site, and calculating an atmospheric quantitative transmission parameter by utilizing the quantitative model; carrying out downsampling processing to an original image which contains fog and haze to obtain a preprocessing template image, and extracting a dark-channel image; extracting the atmospheric optical value A of the original image I(x) which contains the fog and haze, and initializing a depth-of-field correction factor (Omega); optimizing A and Omega in a transmissivity graph of the I(x); estimating a template transmissivity graph t<model> of a preprocessing image, and carrying out upsampling to obtain a transmissivity graph t(x); and according to the t(x) and the I(x), calculating a sharpness processing result image. A fog and haze image sharpness parameter is subjected to closed loop correction through the quantitative model, an image sampling and multi-scale interpolation method is combined to avoid defects that parameters can not be accurately artificially defined and can not be regulated according to practical environment in a dark channel image defogging process, and a sharpness processing effect is improved.

Description

technical field [0001] The invention relates to the technical field of computer image processing, in particular to a method and device for clearing and processing video images in smoggy days. Background technique [0002] When light propagates in media such as fog and haze, the image information collected by the imaging sensor is seriously degraded due to the scattering effect of particles, which limits the application value of the image to a large extent. In order to weaken the impact of haze and other visual obstacles on image quality, especially in areas where computer vision is needed to solve practical problems such as military target recognition, outdoor monitoring and terrain exploration, it is necessary to clear the image. [0003] Currently, there are two common methods in the field of image dehazing: image enhancement-based methods and physical model-based methods. Among them, the dark channel priority algorithm has achieved good results in practical applications ...

AI Technical Summary

Problems solved by technology

[0004] The present invention provides a method and device for clearing video images in foggy days, and performs closed-loop correction on foggy image clearing parameters through a quantitative model, and combines image sampling and multi-scale interpolation methods to avoid artificial definition in the dark channel image defogging process The disadvantages of inaccurate parameters and the inability to adjust according

Images