Defogging method for image containing high-luminance area

Abstract

The invention provides a defogging method for an image containing a high-luminance area. The method comprises the following steps that: S1: selecting an image which contains a high-luminance area andneeds to be subjected to defogging processing; S2: according to saturability, area luminance and the size sequence of RGB (Red, Green and Blue) three channels, segmenting the image into the high-luminance area and a low-luminance area, and independently storing as two picture files; S3: carrying out the defogging processing on the high-luminance area picture file; S4: carrying out the defogging processing on the low-luminance area picture file; and S5: combining a high-luminance area image subjected to the defogging processing and a low-luminance image subjected to the defogging processing, and utilizing an exponential smoothing algorithm to carry out filtering processing on a combined edge area to obtain a final defogging picture. On one hand, the problems that an existing defogging method can not carry out the defogging processing on images under the environments including an artificial light source, a reflection light source, the sky and the like; and on the other hand, the image containing the high-luminance area is subjected to segmentation processing, and the applicable range of the defogging method is enlarged.

Description

technical field [0001] The invention relates to the technical field of image processing and efficiency enhancement, in particular to an image defogging method containing high-brightness regions. Background technique [0002] Image is the foundation of machine vision and the main way and means for the visual system to obtain information. The quality of the image will have a profound impact on the accuracy of subsequent image reading, image analysis, image recognition and other operations. For most vision systems, severe outdoor weather is an important cause of image degradation, among which fog and mist appear most frequently and are widely distributed, which are the main factors affecting the image quality of the vision system. Therefore, there is an urgent need for an efficient and widely applicable image defogging method to meet people's needs. [0003] For the image dehazing problem, many scholars have participated in the research on this problem and proposed their own d...

AI Technical Summary

Problems solved by technology

They believe that the atmospheric light value is the same in the entire image. Such an assumption is not valid for images containing high-brightness areas (sky, artificial light source, reflected light source, etc.), and this assumption will select a larger atmospheric light value. Make the output image overall dark, and even cause problems such as color distortion and blurred details

These methods can only process images of a single depth, and are not ideal for images with multiple distance depths and large amplitude differences. For these complex images, it is necessary to modify the assumption that the overall atmospheric light value remains unchanged.

[0004] These traditional methods can deal with the image defogging problem of a single environment depth, and the scope of application is narrow. However, the actual application environment is complex and changeable, and these methods cannot meet the actual needs. The image defogging method is an inevitable requirement in the field of image defogging

Images