Transformer oil image enhancement method based on gridded gray world white balance algorithm

By gridding the transformer oil image and using the grayscale world white balance algorithm improved by the weighted surround standard deviation, combined with the boundary entropy mechanism, the problem of image color distortion in oil-immersed transformers is solved and the image enhancement effect is improved.

CN117218022BActive Publication Date: 2025-09-09SICHUAN UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202311192748.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-09-15
Publication Date
2025-09-09
Estimated Expiration
2043-09-15

AI Technical Summary

Technical Problem

In existing technologies, in oil-immersed transformers, images captured by robots suffer from color distortion due to changing lighting environments. Traditional grayscale world white balance algorithms perform poorly in scenes with single color blocks or those with limited colors, and cannot effectively eliminate the impact of the lighting environment on color appearance.

Method used

The gridded grayscale world white balance algorithm is adopted to grid the transformer oil image, and the grayscale world white balance algorithm improved by the surround view standard deviation weighting is used for enhancement processing. The gain coefficient is adjusted in combination with the limit entropy mechanism to improve the image quality.

Benefits of technology

The enhancement effect of images under transformer oil is improved, the color distortion problem of traditional methods in single color blocks and scenes with not rich colors is solved, and the image quality is improved without causing overexposure.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117218022B_ABST
    Figure CN117218022B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for enhancing transformer oil images based on a gridded grayscale world white balance algorithm. The method comprises obtaining an original transformer oil image; dividing the obtained original transformer oil image into a grid; and enhancing the original transformer oil image using a grayscale world white balance algorithm improved by weighting the surround standard deviation based on the divided grid image. The present invention solves the problem of poor white balance performance of traditional grayscale world white balance algorithms for single-color objects and scenes with limited color, thereby improving the image enhancement effect of transformer oil images.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Automatic white balance processing method and device for large scene range

    CN110324599A

  • Image processing method and device, electronic equipment and computer readable storage medium

    CN114022375A