Image dehazing method based on adaptive transmittance inpainting based on multi-scale window

Abstract

The invention discloses a multi-scale window-based self-adaptive transmittance repair image defogging method, first obtains a foggy image; uses an 8-direction edge detection operator to perform edge detection on the foggy image, and judges whether it is the edge of the depth of field; according to the discrimination result, automatically Adaptively select the window size to estimate the dark channel; calculate the atmospheric light intensity according to the dark channel estimation map; define the atmospheric dissipation function; obtain the edge-enhanced dark channel estimation map, that is, the atmospheric dissipation value, by guiding the filter; Correct the atmospheric dissipation value to obtain the corrected scattering map (that is, the scattering function value); obtain the projection rate according to the atmospheric light intensity and the scattering function value; bring the projection rate into the dehazing model formula to obtain the image after defogging; The invention effectively reduces the "halo" effect, and the algorithm expands the inapplicability of the dark channel prior principle in areas such as the sky and white objects, and better solves the failure of the dark channel prior principle in bright areas question.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a multi-scale window-based self-adaptive transmittance restoration image defogging method. Background technique [0002] In recent years, my country's domestic smog has continued. However, according to relevant experts' predictions, it will take at least 15-20 years for my country to completely eliminate the impact of smog. Therefore, the research on the clear restoration of foggy images under haze weather has become a major research hotspot in the field of image processing. [0003] At present, the mainstream methods of image defogging at home and abroad can be roughly divided into two categories in principle: the defogging method based on image enhancement and the image restoration method based on physical model. The disadvantage of the former is that the factors that cause image degradation are not considered, and the fog removal is only achieved by incre...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a multi-scale window-based self-adaptive transmittance repair image defogging method to solve the problems existing in the prior art. The present invention can better solve the phenomenon of "halo" effect in the above algorithm and effectively Reduce the time complexity of the algorithm; and solve the problem of the failure of the dark channel prior principle in areas including the sky and white objects, expand the scope of application of the dark channel prior principle, so that it can handle fogged images of various scenes

Images