Face recognition method and system

Abstract

The present invention provides a face recognition method and system, wherein the method includes: extracting the features of all control points of the image to be recognized; taking the control points as bifurcation points, extracting the features of a preset number of control points to form a pre-set tree Set the depth of the decision tree; classify the features of the decision tree to obtain multiple face rectangle images in the image to be recognized; further scale the obtained multiple face rectangle images to a uniform size, and obtain the grayscale image of the face rectangle image ; Input the grayscale image of the rectangular face image into the pre-trained DCNN network to extract the face features of preset dimensions; compare and identify the obtained face features with the face data in the preset database. The invention solves the problem of insufficient computing resources when the traditional high-precision face recognition technology is applied to the mobile platform, and it can still run smoothly under the low-frequency CPU and small memory conditions of the embedded system, while ensuring the accuracy of face recognition .

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a face recognition method and system. Background technique [0002] The research on face recognition technology began in the 1960s, and was improved with the development of computer technology and optical imaging technology after the 1980s, but it really entered the primary application stage in the late 1990s, and the United States, Germany and Japan technology implementation. The key to the success of face recognition technology lies in whether it has cutting-edge core technology to make the recognition result have a practical recognition rate and recognition speed. [0003] However, the current face recognition technology has the following defects: 1. It is sensitive to changes in the face area such as illumination, angle, expression and accessories, resulting in a high recognition rate in an ideal environment, but a poor recognition rate in actual application scenar...

AI Technical Summary

Problems solved by technology

[0003] However, the current face recognition technology has the following defects: 1. It is sensitive to changes in the face area such as illumination, angle, expression and accessories, resulting in a high recognition rate in an ideal environment, but a poor recognition rate in actual application scenarios; 2. The recent DCNN-based face recognition method has high theoretical accuracy, but it cannot run well on mobile platforms with limited computing resources, either because the speed is too slow or the memory is insufficient

Images