Automatic image color matching method and device

Abstract

The invention discloses an automatic image color matching method, medium, equipment and device, and the method comprises the steps: obtaining an original image, and carrying out the saliency detection, so as to determine a foreground region and a background region of the original image; performing color quantization to generate a plurality of first subspaces corresponding to the foreground region and a plurality of second subspaces corresponding to the background region; counting pixel proportions of the first subspace and the second subspace, and averaging pixels; sorting all the first subspaces according to the pixel proportions corresponding to the first subspaces, sorting all the second subspaces according to the pixel proportions corresponding to the second subspaces, and performing secondary clustering on the first subspaces and the second subspaces according to a sorting result; determining a foreground main color and a background main color according to a secondary clustering result, and performing automatic color matching of the original image according to the foreground main color and the background main color; the main color conforming to human eye perception in the original image can be effectively and automatically extracted, and then the final automatic color matching effect of the image is improved.

Description

technical field [0001] The invention relates to the technical field of image color matching, in particular to an image automatic color matching method, a computer-readable storage medium, a computer device and an image automatic color matching device. Background technique [0002] In the poster intelligent design system, it is necessary to extract the main color of the picture uploaded by the user, so as to determine the existing color scheme according to the extraction result. [0003] However, most of the existing dominant color extraction algorithms only determine the final dominant color based on the overall proportion of the color in the picture. It can be understood that the color with a large overall proportion in the picture is not necessarily the main color of the picture (for example, a little red in the green bushes, although the proportion of red is relatively large at this time, it is not the main color, and the main color should be a small proportion. the red ...

AI Technical Summary

Problems solved by technology

Therefore, the main color selected by the existing main color extraction algorithm is often contrary to the perception result of the human eye; secondly, certain professional experience is required to artificially color the image, and the corresponding color matching result of a single user on the image is generally monotonous. The richness is not enough to adapt to the aesthetics of different users on images

Images