Stereoscopic face tracking and expression identification method

Abstract

The invention provides a stereoscopic face tracking and expression identification method. The method includes S1, image preprocessing, S2, face detection, S3, expression identification and S4, face tracking. Image preprocessing of Step S1 includes S11, image graying, S12, image equalization and S13, image binarization operation, face detection of Step S2 uses Harr features and an AdaBoost classifier to perform face detection, expression identification of Step S3 uses wavelet transformation and a nearest neighbor classifier to realize expression identification, and face tracking of Step S4 usesa feature matching method to calculate parallax of left and right camera images, calculate a three-dimensional coordinate of the face, and calculate to obtain an angle by which a holder needs to rotate, thereby realizing tracking of a face region.

Description

technical field [0001] The invention relates to the field of human-computer interaction, in particular to a three-dimensional human face tracking and expression recognition method. Background technique [0002] Facial expression not only plays an important role in human-to-human communication, but also is an important research content to enable computers to understand the intention expressed by human expressions more accurately and realize human-computer interaction. [0003] Most of the existing face tracking and expression recognition systems are aimed at videos from fixed cameras, using face detection technology and expression recognition technology for face tracking and recognition. However, this type of system is limited by the angle and position of the camera and cannot Acquiring positive facial expression images or videos, resulting in recognition errors. In addition, face detection systems often use a single camera for tracking. It is difficult to locate the three-d...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to overcome the deficiencies of the prior art, especially to solve the existing face tracking and expression recognition system, limited by the angle and position of the camera, it is impossible to obtain the front face expression image or video, resulting in recognition errors; and Using a single camera for tracking, it is difficult to achieve three-dimensional position positioning of the face area with a single camera, and can only achieve positioning in a two-dimensional plane. For people walking from a distance, it is easy to lose the tracking area.

Images