An image processing method and device based on prior knowledge

Abstract

The invention provides an image processing method and device based on prior knowledge. The image processing method based on prior knowledge comprises the steps that the first skin color region of theimage of a chromaticity space is acquired; according to the chromaticity information of each pixel in the first skin color region and a preset skin color range, the reliability value of the first skincolor region is determined; if the reliability value of the first skin color region is not greater than a preset threshold, the position of each pixel in the first skin color region is adjusted according to the position of each pixel in the first skin color region and the preset skin color range to acquire a second skin color region, wherein the relative positions of pixels in the second skin color region correspond to the relative positions of pixels in the first skin color region; the adjustment value of the image in the chromaticity space is determined according to the first skin color region and the second skin color region; and image color cast adjustment is carried out according to the adjustment value of the image in the chromaticity space to acquire an image after white balance. Simple and efficient image aberration reduction is realized. The correction effect of the image is improved.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to an image processing method and device based on prior knowledge. Background technique [0002] Auto White Balance (Auto White Balance) is a very important link in the digital image preprocessing system. The so-called white balance can be understood as restoring the photo color of a white object to white. Objects have their own colors because their surfaces selectively absorb and reflect visible light in part of the wavelength range, and the reflected light is the color of the object. When the light of the light source is not natural light but has a certain color, the light reflected by the object is not only its own color, but also has the color of the light source. The human eye has a strong ability to adapt, and can automatically shield part of the color of the light source in an environment with colored light sources, that is, from the perspective of human pe...

AI Technical Summary

Problems solved by technology

However, because the grayscale world algorithm only directly weights the RGB components of the pixels, it cannot process the image more intelligently. Although it can correct the white restoration to a certain extent, it loses the brightness of the image, and it is not rich in single color. poor image correction

The total reflection theory algorithm is not effective for image processing with severe color cast. Since the theory is based on the area with the largest brightness in the image, the correction algorithm will produce errors for images where the area with the largest brightness in the image is not a white area.

Images