Classroom attendance method and system based on face recognition

Abstract

The invention discloses a classroom attendance method and system based on face recognition. The method comprises the following steps of S1, continuously acquiring video images by a camera and transmitting the video images to a server; S2, carrying out face alignment and coordinate information acquisition on classroom image frames acquired in the S1 by using a face detection algorithm; and S3, aiming at each image frame, carrying out feature point extraction on a detected face, comparing extracted eigenvectors with face features in a database, and judging whether the detected face is the same as faces in the database or not. The method and the system have the beneficial effects that two-dimensional probability distribution of each face in the plane is calculated through a face recognition result of the multi-frame image, a position distribution probability maximum point of the face is obtained, and a final attendance result is determined through threshold screening. Compared with the single-frame image, the method has the advantage that the accuracy of the recognition result is higher, so that the judgment error of the student attendance result due to the recognition error of the single-frame image by the system is avoided.

Description

technical field [0001] The invention relates to the field of teaching technology, in particular to a face recognition-based classroom attendance method and system. Background technique [0002] In the smart classroom attendance system based on face recognition currently used by schools, there are two methods: single-frame image face recognition and multi-frame image face recognition. However, there are certain deficiencies in the two existing methods. When using single-frame image face recognition to determine the attendance results of students, the system is easy to identify errors, resulting in students not coming and mistakenly recognized as present or students coming and recognized as absent. In the case of continuous tracking when using multi-frame image face recognition, it is prone to crossover errors. It can be seen that the misjudgment rate of the smart classroom attendance system in the current school is high, and the accuracy rate of the system's face recognition...

AI Technical Summary

Problems solved by technology

However, there are certain deficiencies in the two existing methods. When using single-frame image face recognition to determine the attendance results of students, the system is easy to identify errors, resulting in students not coming and mistakenly recognized as present or students coming and recognized as absent. In the case of continuous tracking when using multi-frame image face recognition, it is prone to crossover errors

It can be seen that the misjudgment rate of the smart classroom attendance system in the current school is high, and the accuracy rate of the system's face recognition is low.

Images