An image adjustment method for uniform skin color

Abstract

The disclosure relates to the technical field of data imaging processing, and discloses an image adjustment method for uniform skin color, which mainly includes generating a face skin mask, counting face skin information, and performing skin correction according to the statistical information. By combining skin color detection, face key point detection, and skin statistical information to generate accurate face red and yellow patch mask maps, it can provide accurate guidance for the uniform skin color. Adjusting the hue channel separately retains the brightness of the original image, and the result is more natural, and only processes areas within the range of 280°-260° and 0°-60°, which can more accurately process reddish and yellowish skin tones and achieve redness removal. The effect of skin blemishes such as bloodshot, plateau red, erythema and macula.

Description

technical field [0001] The disclosure belongs to the technical field of data imaging processing, and in particular relates to an image adjustment method for uniform skin color. Background technique [0002] In recent years, digital photography devices such as digital cameras, smart phones, and tablet computers have become mainstream tools for people to take pictures. People can easily use these devices to record and share their lives through photos, and many of them are photos with people as the main body. , and the location and environment of taking pictures are very different. The photos taken are usually affected by the environment or the difference in equipment. The effect of the picture is always unsatisfactory. People care about the skin color in the image. There are large differences in the skin color of different skin areas of people, which affects people's image temperament, especially when there are red blood streaks, plateau red and macula on the face, the imaging...

AI Technical Summary

Problems solved by technology

[0002] In recent years, digital photography devices such as digital cameras, smart phones, and tablet computers have become mainstream tools for people to take pictures. People can easily use these devices to record and share their lives through photos, and many of them are photos with people as the main body. , and the location and environment of taking pictures are very different. The photos taken are usually affected by the environment or the difference in equipment. The effect of the picture is always unsatisfactory. People care about the skin color in the image. There are large differences in the skin color of different skin areas of people, which affects people's image temperament, especially when there are red blood streaks, plateau red and macula on the face, the imaging of the corresponding part of the area will appear excessive red or yellow, and the image display is not beautiful , so the processing of skin color before imaging becomes an important link

[0003] However, when the current automatic image processing technology processes the skin color, it is mostly processed according to the overall color information contained in the image, and the area outside the skin is often adjusted uniformly. Compared with the normal imaging, there is a large color difference, which makes the imaging effect extremely unnatural

For example, when the image is in RGB mode and the content of the color is adjusted, the color of non-target areas such as hair, facial features, and clothing areas will be adjusted together, resulting in discoloration; when the image is in HSV mode and the saturation is adjusted, it will be Make non-target areas appear too dark, too light, etc., reducing the quality of the image

Images