A skin color detection method, device and terminal

Abstract

Embodiments of the present invention provide a skin color detection method, device, and terminal, wherein the method includes: obtaining a picture to be processed, performing face recognition on the picture to be processed; if a face is recognized, obtaining the person in the picture to be processed Face area; According to the first skin color lookup table of the first template picture set, determine the skin color lookup table of this people's face area; According to the skin color lookup table of this people's face area and the second skin color lookup table of the second template picture set, determine the skin color lookup table The skin color lookup table of the picture to be processed; the skin color detection of the picture to be processed is performed by using the skin color lookup table of the picture to be processed. Through the embodiments of the present invention, the accuracy of skin color detection for pictures can be improved.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a skin color detection method, device and terminal. Background technique [0002] At present, more and more smart phones, tablet computers and other terminals have built-in beautification functions, so that when users use the terminal to take selfies or take pictures of others, they can add skin care, whitening and other beautification to the faces in the pictures. effect to get a character picture with better visual effect. To add beautification effects to human faces, it is first necessary to identify the face area in the picture, and then identify the skin-colored area in the face area. Generally, it is only necessary to add a beautifying effect to the skin-colored area, while retaining non-skinned areas (such as eyes, eyebrows, etc.) etc.) without embellishing its authenticity. [0003] The way of the existing skin color detection scheme is: identify the hu...

AI Technical Summary

Problems solved by technology

However, this scheme will cause false detection of skin color when glasses or other objects are worn on the eyes. The ellipse model is a parameterized skin color model obtained from a large amount of training data. The distribution of skin color in a specific frame is not necessarily an ellipse. The skin color model's skin color detection for specific frames may also cause false detections. In addition, missed detections of face detection (such as missed detections of side faces or frontal faces) will cause missed detections of skin colors

It can be seen that the above scheme is less accurate in skin color detection

Images