Image correction method, image correction device and electronic equipment

Abstract

The invention discloses an image correction method, an image correction device and electronic equipment. According to one embodiment, the image correction method can comprise the steps: receiving a to-be-corrected image; and performing color correction on the to-be-corrected image by using a color correction model, wherein the color correction model is obtained by training images of color cards atdifferent color temperatures. In some examples, the image of the color card may be an original RAW image. The color correction model is trained by using the original RAW images of the color card under different color temperatures, so the influence of factors such as ambient light can be eliminated, high correction accuracy is realized, and the color correction method has a wide application range.

Description

technical field [0001] The present application relates to the field of image processing, and more specifically, relates to an image correction method, an image correction device and electronic equipment. Background technique [0002] When the user takes an image with a camera, etc., he expects the image to reflect the real scene. However, due to the influence of factors such as ambient light and the camera's own exposure, the original image signal obtained by the camera often has certain errors compared with the real scene. For example, the ambient light in the scene may cause a color cast in the image captured by the camera. For example, the image captured under fluorescent light is greenish, while the image captured under tungsten light is yellowish. Therefore, in order to solve this problem, image color correction is widely used in various cameras. [0003] Existing image color correction methods include linear fitting, polynomial fitting, and neural network color correc...

AI Technical Summary

Problems solved by technology

[0003] Existing image color correction methods include linear fitting, polynomial fitting, and neural network color correction, etc., which generally use RGB data of several sampling points for correction. Since RGB data is preprocessed data, it is used as network input. There is a certain error, and the existing correction method does not eliminate the interference of ambient light

In addition, the regression accuracy of traditional linear fitting or polynomial fitting is not ideal, and the lack of data leads to poor generalization ability of the model, so it cannot express the regression effect of the image well

Images