Face identification system

Abstract

The invention discloses a face identification system, including hardware equipment, a software engine and a database. The hardware equipment comprises a computer, a switch and a video input device. The software engine comprises a face capture engine, an interference reduction engine, a face modeling engine and a face comparison engine. The database stores a photograph database and a feature value template database. The computer is connected to the video input device through the switch. The photograph database stores original comparison photographs. The face comparison engine collects and obtains 2D human images including face biological features in video or image information through the video input device. The interference reduction engine corrects and restores the captured 2D human images, and removes human images which do not accord with modeling standards. The face modeling engine generates 3D feature values through the collected 2D human images which accord with the modeling standards, and stores the 3D feature values in the feature value template database. The face comparison engine compares the 3D feature values generated by the 2D human images with 3D feature values generated by photographs in the photograph database to output results.

Description

technical field [0001] The invention relates to a computer-aided biological feature recognition system, more specifically a face recognition system. Background technique [0002] There are many methods of face recognition, but they can be roughly divided into methods based on color block analysis and comparison and methods based on face template analysis and comparison. [0003] The method based on color block analysis and comparison has high requirements on the environment and background, as well as the sensitivity to light, so it is subject to many interference factors, resulting in a low recognition rate. [0004] The comparative analysis method based on 2D face models also faces two difficult problems: [0005] The first is the position and angle of the face. Because of the dynamic acquisition, the face in the photo is different from the static face. The posture of the person in reality is random, and the angle deviation in the left, right, up and down, and even the dir...

AI Technical Summary

Problems solved by technology

[0003] The method based on color block analysis and comparison has high requirements on the environment and backgro

Images