Transformer substation video image sharpening method in accordance with human eye visual characteristics

Abstract

The present invention discloses a transformer substation video image sharpening method in accordance with human eye visual characteristics, and belongs to the field of image communication, video image processing, intelligent power station and the like. According to the invention, the method comprises the steps of 1) acquiring the video image of a transformer substation; 2) based on the partitioning principle, adaptively acquiring an atmospheric brightness value; 3) according to an atmospheric optical imaging model, establishing an image degradation equation and solving out a rough transmittance and other parameters based on the dark primary color priori knowledge; 4) with the gray-level image of an original image as a bootstrap image, guiding a rough atmospheric light curtain in combination with the bilateral filtering means so as to obtain a new atmospheric light curtain, and further reversely deducing a preliminary restored image; 5) in combination with human eye LMS visual characteristics, enhancing the restored image under an LIPC color logarithmic image processing framework so as to obtain a fog-free clear image more consistent with the habits of human eyes. The method is remarkable in defogging effect, clear in details, natural in color, and wide in application prospect. The method effectively solves the video definition problem of transformer substations in the prior art.

Description

technical field [0001] The invention relates to the technical fields of image communication, video image processing, and intelligent power station, and more specifically, relates to a substation video image clearing method that conforms to the visual characteristics of human eyes. Background technique [0002] Substations and power facilities are spread all over the place, and most of them are unattended stations, which need to monitor the real-time situation through surveillance cameras. However, due to the scattering effect of atmospheric particles in foggy days, the images captured by the surveillance cameras in substations are blurred, and it is difficult to distinguish people, animals, and unknown sundries that violate regulations. And hidden dangers cannot be dealt with in advance. Although infrared cameras are installed in some key areas, they are expensive and cannot be densely deployed, making it difficult to implement them across the entire network. [0003] In r...

AI Technical Summary

Problems solved by technology

According to the atmospheric imaging model, the present invention adaptively obtains the atmospheric brightness in combination with the block idea, establishes the image degradation equation by using the prior knowledge of the dark channel, and proposes a more accurate transmission rate estimation method based on guided joint bilateral filtering, in order to To further improve the appreciation of the restored image, the present invention combines the visual characteristics of the human eye to enhance the color of the restored image under the framework of LIPC (Color Logarithmic Image Processing), effectively solving the problem of poor visibility of the substation monitoring video in foggy days

Images