Method and device for recognizing faces through multi-angle imaging

Abstract

The invention discloses a method and a device for recognizing faces through multi-angle imaging. The method comprises the steps that optical devices are installed by an included angle of 100 degrees-170 degrees according to preset position and range; cameras are placed at the opposite positions of the optical reflection devices so as to simultaneously collect the face images of a person to be shot appearing in each optical reflection device; multiple face images of the person to be shot forming at different angles are obtained; the face images of different angels are respectively sent to a face recognition module for classification and recognition; the recognition results of all the face images are screened, and the most dependable image is selected as the final face recognition result. According to the method, the optical reflection device is used for simultaneously capturing the face images of one person at different angles for recognizing the faces, effective capture of face images of multiple angles is beneficial for the system to get the optimal recognition result, and the system is difficult to be affected by the change of face gestures, so that the recognition efficiency and the adaptability are high.

Description

technical field [0001] The invention belongs to the technical fields of pattern recognition, image processing and computer vision, and in particular relates to a method and device for face recognition by taking images from multiple perspectives. Background technique [0002] Face recognition is a kind of biological feature recognition technology. Under ideal experimental conditions, face recognition has achieved good results, and the accuracy rate can reach more than 95%. However, in the actual application environment, the face recognition The attitude is uncertain, and it is difficult to ensure that the camera can capture the face image from the same angle every time, which leads to a sharp drop in the accuracy of face recognition. One solution is to use multiple cameras from different angles of view. imaging and identification. [0003] It is inconvenient to use multiple cameras at the same time, and the viewing angles of the cameras need to be adjusted separately. The ca...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the embodiments of the present invention is to provide a method and device for multi-view image acquisition face recognition, which aims to solve the problem of using multiple cameras at the same time and need to adjust the viewing angles of the cameras separately. The cameras are easy to interfere with each other, and it is difficult to ensure simultaneous completion The collection of face images cannot be accurately synchronized, which increases the complexity and cost of the system, increases the possibility of failure, and is not easy to maintain

Images