Foggy image clarity method based on adaptive clustering color transfer

Abstract

The invention discloses a fog image clearing method based on adaptive clustering color transfer. The method is implemented according to the following steps: collecting source image and target image information, respectively performing decoupling processing through color space conversion; The mean and variance of the image and the target image make the statistical characteristics of the source image "closer" to the target image as much as possible; the obtained corrected source image in the Lαβ color space is converted from the Lαβ color space to the RGB color space, Obtain a color transfer correction result map; continue to perform color clustering on the result image after the first color transfer correction and the target image; search and correspond to similar categories; perform secondary color transfer correction; manually adjust the clustering of the secondary correction to obtain the final calibration results. The method of the invention realizes the clear processing of the image taken under the foggy condition, and can restore the effective information in the source image.

Description

technical field [0001] The invention belongs to the technical field of image restoration, and relates to a method for clearing and restoring images under low-contrast and low-information conditions, and in particular to a foggy image clearing method based on adaptive clustering color transfer. Background technique [0002] With the continuous development of computer image processing technology, as well as the urgent needs of monitoring, digital photography and other fields, when people analyze some low-contrast and low-information photos, they need to be able to clear the images to a greater extent and restore them. Some key information in the image. [0003] In my country, heavy fog is a frequent weather condition. In heavy fog, no matter which field of outdoor video surveillance system is used in, all the images captured are low-contrast images with heavy fog interference. The amount of image information is very low, and there are a lot of hard-to-see picture information ...

AI Technical Summary

Problems solved by technology

The amount of image information is very low, and there are a lot of hard-to-see picture information and a considerable part of the detail information that cannot be distinguished at all, and these are sometimes the most needed content of an image. Therefore, under normal circumstances, heavy fog The utilization rate of the following images is quite low, and even in some cases, they have to be treated as waste images without any value of use and analysis

[0004] Under normal circumstances, the method of stretching the contrast is used to sharpen low-contrast images. However, due to the inconsistency of the fog conditions in the actual scene, the degree of stretching required is also very different, and it is difficult Find the mapping relationship between fog conditions and contrast, so in many cases, even if the contrast is manually stretched nonlinearly, it is difficult to achieve better results, which makes image restoration in foggy weather become a tricky question

Images