Human face identifying system and method applied to security robot

Abstract

The invention relates to a human face identifying system and a method applied to a security robot. In response to the problems that the fixed gate access control type human-face identifying system has a single environment and that its human face characteristics description ability is insufficient, the invention provides this moving human face identifying system based on a mobile robot and deep learning. With regard to the equipment, the system realizes the human face identifying under the conditions that either a person or the robot moves or both the person and the robot moves at the same time. In addition, through the use of an infrared camera, human face identifying in complete darkness at night is realized. With regard to the algorithm, the system serves as a complete system from the human face detection and characteristic point detection to the human face correcting and human face identifying. Through the decrease in the number of the characteristic points in the characteristic point detection module and through the utilization of a deep convolution neural network model by the human face identifying module, it is possible to realize the rapid identifying of the human face at high accuracy.

Description

technical field [0001] The invention relates to the field of face recognition, in particular to a face recognition system and method applied to security robots. Background technique [0002] The face is one of the most important biometric features of the human body. Relying on the face for identification has become an important means of identifying wanted criminals and other specific persons. However, in the current era of big data, it is time-consuming and laborious to retrieve data manually. Not feasible. Existing automatic face recognition technology can only be applied to fixed-position frontal face recognition, such as common access control systems, and the recognition rate is very low in complex dynamic environments. Traditional face recognition technology uses artificially designed face image features such as SIFT, and then uses linear methods to extract feature representations on the face level, such as using PCA or LDA to obtain face features such as Eigenface or F...

AI Technical Summary

Problems solved by technology

[0008] Aiming at the problems that the current fixed access control face recognition system has a single environment and insufficient line

Images