Face recognition method and system for binocular camera

Abstract

The invention discloses a face recognition method and system for a binocular camera, and relates to the field of face feature recognition. The method comprises the following steps: S01, a left cameraand a right camera of an eye camera simultaneously complete face two-dimensional image acquisition; S02, performing two-dimensional feature on the two-dimensional image by using an active shape modeltechnology in a processor; S03, obtaining three-dimensional coordinates of the feature points by combining camera calibration parameters; S04, sending the calibrated three-dimensional coordinates to acomparison data platform; s05, performing identification by using a data platform neural network; s06, feeding back an identification result to the processor; and S07, the processor controlling the equipment to execute related operation. According to the method, the left camera and the right camera in the binocular cameras are used for collecting different face images, two-dimensional feature point calibration is carried out, calibrated parameters are used for constructing three-dimensional features, the three-dimensional features are input into the neural network for recognition, and compared with existing two-dimensional recognition, the recognition accuracy is improved, and the situation that pictures are used for deception is avoided.

Description

technical field [0001] The invention belongs to the field of face feature recognition, and in particular relates to a face recognition method of a binocular camera and a face recognition system of a binocular camera. Background technique [0002] At present, face recognition technology is widely used in TVs and various electronic display products. Due to its convenient and unique biometric performance, it has an indispensable place in the application of future TV functions. In particular, as a household appliance such as a TV, the family members it faces have some differences in the application environment of ordinary face recognition, that is, it is very easy for family members to obtain mutual personal information, such as photos of faces. , so that face recognition can be deceived through effective photos, making the application of face recognition on these products flawed. [0003] The application of existing face recognition technology on TV and various electronic dis...

AI Technical Summary

Problems solved by technology

[0004] The object of the present invention is to provide a face recognition method and system of a binocular camera, collect different face images through the left and right cameras in the binocular camera, perform two-dimensional feature point calibration, and use the calibrated parameters to construct a three-dimensional feature input to Recognition by neural network solves the problem that the existing face recognition adopts two-dimensional recognition, it is easy to deceive through photos, and the recognition accuracy is insufficient

Images