Method for optimizing white balance of face area of camera

Abstract

The invention discloses a method for optimizing white balance of a face area of a camera. The existing method does not aim at the whole image and is more prone to global correction, so that color cast of a human face is easily caused. According to the method, an image collected by a CMOS sensor is sent to an image processor, a processed image is obtained, and face area information is detected through a face detection network; extracting a face image according to the face region information, and splitting the face image into a left face image and a right face image; performing histogram statistics on the left and right images of the human face, judging a light environment of a human face region, and obtaining a white balance weight of the human face region; and mixing the white balance weight of the face region with the original weight of the system to obtain a final white balance weight, and setting the final white balance weight in the system. According to the method, the neural network technology is combined, the influence of complex environment scenes on the face color of the camera is fully considered, the color of the face area can be adaptively corrected under different environment light and scenes, and the white balance effect of the face area of the camera is improved.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a method for optimizing the white balance of a face area of ​​a camera. Background technique [0002] Since the output of the CMOS sensor is a Bayer diagram, the color itself is in an unbalanced state, so the color of the output image is not the color presented in nature, which causes people to see that the image is very strange, and solving the problem of image color becomes an unavoidable problem. The problem. At present, in the image processing of the camera, automatic white balance is performed to correct the color. The traditional white balance methods are: the whole image is analyzed by global statistics or local white point finding, and the image is globally corrected by the theory of gray world. Because the details in the image are more complex and there are many contents in the image, the method of global color correction cannot guarantee the corre...

AI Technical Summary

Problems solved by technology

Since the details in the image are more complex and there are many contents in the image, the method of global color correction cannot guarantee that every color is corrected correctly; and the content that the observer pays attention to is different. For the face camera, The main body of content that obser

Images