CN117541442A

CN117541442A - Teaching attendance management method, device, equipment and storage medium

Info

Publication number: CN117541442A
Application number: CN202311426880.4A
Authority: CN
Inventors: 王东; 徐宝林; 李春平
Original assignee: Guangdong Baiyun University
Current assignee: Guangdong Baiyun University
Priority date: 2023-10-30
Filing date: 2023-10-30
Publication date: 2024-02-09

Abstract

The invention discloses a teaching attendance management method, a teaching attendance management device, teaching attendance management equipment and a storage medium, and belongs to the technical field of computer vision. The method comprises the following steps: acquiring a classroom video stream through a monitor; based on a preset student face feature library, carrying out face recognition according to the classroom video stream to obtain a face recognition result set; based on a preset attendance management algorithm, generating an attendance management result according to the face recognition result set. Through the face feature library and the teaching attendance management algorithm, accurate teaching attendance management is realized. Therefore, intelligent attendance management is realized, and the technical problems of low teaching attendance management efficiency and low precision in the prior art are solved. Compared with the prior art, the method has the advantages of short time consumption and high efficiency.

Description

Teaching attendance management method, device, equipment and storage medium

Technical Field

The present invention relates to the field of computer vision, and in particular, to a method, apparatus, device, and storage medium for teaching attendance management.

Background

In the teaching process of the current colleges, attendance is a very necessary and mandatory tool in the study, helps a lecturer to independently track students and measure their interests, is one of key factors for determining the performance of the students, and is positively correlated even though the high attendance rate and good learning performance of the students are not necessarily related, the students with good performance tend to have higher attendance rate, and the attendance management is helpful for improving the learning performance of the students and perfecting the teaching system of the colleges.

At present, the attendance checking mechanisms and equipment adopted by part of universities are relatively backward and have quite a lot of loopholes, the traditional paper particle rosters are troublesome and laborious, especially when the number of people is more than hundred people in class, and the attendance checking efficiency is quite low; some existing equipment attendance checking methods have the problem that positioning accuracy and recognition accuracy are not accurate enough.

Disclosure of Invention

The invention mainly aims to provide a teaching attendance management method, a device, equipment and a storage medium, which aim to solve the problems of low teaching attendance management efficiency and low precision.

In order to achieve the above purpose, the present invention provides a teaching attendance management method, which includes the following steps:

acquiring a classroom video stream through a monitor;

based on a preset student face feature library, carrying out face recognition according to the classroom video stream to obtain a face recognition result set;

based on a preset attendance management algorithm, generating an attendance management result according to the face recognition result set.

Optionally, the step of obtaining the face recognition result set based on the preset student face feature library and performing face recognition according to the classroom video stream includes:

s01: reading the frame number of the classroom video stream to obtain a classroom picture;

S02: preprocessing the classroom picture to obtain a preprocessed classroom picture;

s03: carrying out face capturing on the preprocessed classroom picture through a preset face capturing algorithm to obtain face data to be recognized;

s04: based on a preset prediction function, carrying out face recognition on the face data to be recognized according to the face feature library of the student by a preset face recognition device to obtain a face recognition result;

repeating the steps S01-S04 until the preset timing task is finished, and obtaining a plurality of face recognition results;

and encoding the face recognition results based on a preset time sequence encoding algorithm, and establishing the face recognition result set.

Optionally, the step of generating the attendance management result based on the preset attendance management algorithm according to the face recognition result set includes:

based on a classroom management algorithm in the attendance management algorithm, generating a classroom management result according to the face recognition result set;

generating a student fatigue detection result set according to the face recognition result set based on a fatigue detection algorithm in the attendance management algorithm, wherein the fatigue detection algorithm is constructed based on a gray level detection technology;

And obtaining the attendance management result according to the classroom management result and the student fatigue detection result set.

Optionally, the step of generating a classroom management result according to the face recognition result set based on the classroom management algorithm in the attendance management algorithm includes:

classifying the face recognition results in the face recognition result set based on a preset time sequence classification algorithm to obtain an intra-class face recognition result set, a pre-class face recognition result and a post-class face recognition result;

based on a check-in management algorithm in the classroom management algorithm, generating a check-in result according to the pre-class face recognition result;

based on a attendance quality detection algorithm in the classroom management algorithm, generating an attendance quality detection result set according to the intra-class face recognition result set;

based on a sign-in management algorithm in the classroom management algorithm, generating a sign-in result according to a post-class face recognition result;

and obtaining the classroom management result according to the check-in result, the attendance quality detection result set and the check-out result.

Optionally, the step of generating the attendance quality detection result set according to the intra-class face recognition result set based on the attendance quality detection algorithm in the classroom management algorithm includes:

Acquiring a first intra-class face recognition result in the intra-class face recognition result set;

s05: judging whether the intra-class face recognition result is consistent with the last face recognition result;

s06: if the in-class face recognition result is inconsistent with the last face recognition result, generating a corresponding attendance quality detection result according to the in-class face recognition result based on the attendance quality detection algorithm;

s07: acquiring the next intra-class face recognition result according to the time sequence;

repeating the steps S05-S07 until the intra-class face recognition result set is traversed, and establishing the attendance quality detection result set.

Optionally, the step of generating the student fatigue detection result set according to the face recognition result set based on the fatigue detection algorithm in the attendance management algorithm includes:

acquiring a first face recognition result in the face recognition result set;

s08: judging whether the face recognition result is matched with the previous face recognition result or not;

s09: if the face recognition result is not matched with the previous face recognition result, analyzing the low head condition of the student in the face recognition result based on the fatigue detection algorithm to obtain a student fatigue detection result;

S10: acquiring the next face recognition result according to the time sequence;

repeating the steps S08-S10 until the face recognition result set is traversed, and establishing the student fatigue degree detection result set.

Optionally, the step of generating the attendance management result based on the preset attendance management algorithm according to the face recognition result set further includes:

and generating an attendance report according to the attendance management result based on a preset big data processing algorithm.

In addition, in order to achieve the above object, the present invention also provides a teaching attendance management device, which includes:

the video acquisition module is used for acquiring a classroom video stream through the monitor;

the face recognition module is used for carrying out face recognition according to the classroom video stream based on a preset student face feature library to obtain a face recognition result set;

and the attendance management module is used for generating an attendance management result according to the face recognition result set based on a preset attendance management algorithm.

The embodiment of the invention also provides a terminal device which comprises a memory, a processor and a teaching attendance management program which is stored in the memory and can run on the processor, wherein the teaching attendance management program realizes the steps of the teaching attendance management method when being executed by the processor.

The embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium is stored with a teaching attendance management program, and the teaching attendance management program realizes the steps of the teaching attendance management method when being executed by a processor.

The embodiment of the invention provides a teaching attendance management method, a device, equipment and a storage medium, wherein a classroom video stream is acquired through a monitor; based on a preset student face feature library, carrying out face recognition according to the classroom video stream to obtain a face recognition result set; based on a preset attendance management algorithm, generating an attendance management result according to the face recognition result set. Through the face feature library and the teaching attendance management algorithm, accurate teaching attendance management is realized. Therefore, intelligent attendance management is realized, and the technical problems of low teaching attendance management efficiency and low precision in the prior art are solved. Compared with the prior art, the method has the advantages of short time consumption and high efficiency.

The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.

Drawings

FIG. 1 is a schematic diagram of functional modules of a terminal device to which the teaching attendance management device of the present invention belongs;

FIG. 2 is a flow chart of a first exemplary embodiment of a teaching attendance management method according to the present invention;

FIG. 3 is a schematic diagram of the teaching attendance management algorithm in the teaching attendance management method according to the present invention;

FIG. 4 is a flow chart of a second exemplary embodiment of a teaching attendance management method according to the present invention;

FIG. 5 is a schematic flow chart of a face feature library constructed in the teaching attendance management method of the invention;

FIG. 6 is a flow chart of a third exemplary embodiment of a teaching attendance management method according to the present invention;

FIG. 7 is a flowchart of a fourth exemplary embodiment of a teaching attendance management method according to the present invention;

FIG. 8 is a functional block diagram of an attendance system in the teaching attendance management method of the present invention;

FIG. 9 is a flowchart of a fifth exemplary embodiment of a teaching attendance management method according to the present invention;

FIG. 10 is a flowchart of a sixth exemplary embodiment of a teaching attendance management method according to the present invention;

fig. 11 is a flowchart of a seventh exemplary embodiment of the teaching attendance management method according to the present invention.

The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.

Detailed Description

It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.

The main solutions of the embodiments of the present invention are: acquiring a classroom video stream through a monitor; based on a preset student face feature library, carrying out face recognition according to the classroom video stream to obtain a face recognition result set; based on a preset attendance management algorithm, generating an attendance management result according to the face recognition result set. Through the face feature library and the teaching attendance management algorithm, accurate teaching attendance management is realized. Therefore, intelligent attendance management is realized, and the technical problems of low teaching attendance management efficiency and low precision in the prior art are solved. Compared with the prior art, the method has the advantages of short time consumption and high efficiency.

Technical terms related to the embodiment of the invention:

face recognition: the face recognition technology is widely applied and is based on facial features of people, the identity of the input face image is recognized, and the identity of the person corresponding to the extracted face information is recognized by detecting the face information in the image.

haarcascade_front face_default. Xml file:

haarcascade_front face_default. Xml is a cascade classifier file provided by OpenCV for face detection. The cascade classifier is an object detection method based on machine learning, and the classifier obtained through training can be used for detecting the human face in the image.

The XML file contains configuration and parameter information of a cascade classifier for describing a face detection algorithm. The cascade classifier consists of a plurality of weak classifiers, each using Haar features to determine if a face is present. Haar features are obtained by calculating pixel differences between different regions of the gray scale image.

After the OpenCV function is used to load the haarcascade_front face_default. Xml file, the haarcascade_front face_default. Xml file can be applied to an image or video stream to realize face detection. The detected face is marked and can be used for the tasks of face recognition, expression analysis, face feature extraction and the like.

The haarcascade_front face_default. Xml file is used only for face detection, not to identify a particular face. To perform face recognition, more complex algorithms and training models are typically required.

In summary, haarcascade_front face_default. Xml is a cascade classifier file for face detection, which can be used to detect faces in images or videos, and mark bounding boxes around the detected faces.

Haar classifier: haar classifier is a cascade of classifiers for object detection and recognition, originally proposed by Viola and Jones in 2001, and is widely used in the fields of image processing and computer vision.

Haar classifiers are based on Haar features (Haar-like features) that represent the features of a target object by computing differences in pixel values for different regions in the image. Haar features may describe local features of edges, line segments, corners, etc. of an object.

A Haar classifier is a multi-layered cascade of classifiers consisting of many weak classifiers. Each weak classifier uses Haar features to distinguish between target objects and background. The cascading features enable the Haar classifier to have efficient performance when processing a large number of images.

The training process of the Haar classifier is completed through a machine learning method. In the training process, a positive sample (including an image of a target object) and a negative sample (including an image of a background only) are used for training, and a weak classifier is selected and optimized through an iterative process, so that a high-efficiency accurate cascade classifier model is finally obtained.

In practical applications, haar classifiers are often used for face detection, pedestrian detection, and other tasks. For example, in face detection, a trained Haar classifier model may be used to detect faces in images or videos and to mark the detected face locations.

In summary, a Haar classifier is a cascade of classifiers based on Haar features for object detection and recognition. The cascade model formed by a plurality of weak classifiers is obtained through training, and the target object in the image can be efficiently and accurately detected.

LBPH identifier: LBPH (Local Binary Patterns Histograms) identifier is a face recognition method based on local binary pattern histogram. It is widely used in the field of computer vision, especially in face recognition tasks.

The LBPH identifier represents the face features by extracting a local binary pattern (Local Binary Patterns, abbreviated as LBP) and histogram statistics on the face image. LBP is a method for describing texture features that compares each pixel in an image with its neighbors and encodes the comparison result into a binary number.

In the LBPH recognizer, the face image is first divided into different small regions, and then LBP feature extraction is performed on each region. For each pixel, the LBP value at that pixel location is 1 if the gray value of the adjacent pixel is greater than or equal to the gray value of the center pixel, and is otherwise 0, as compared to its adjacent pixels. Through such comparison and encoding processes, LBP features of the entire face image can be obtained.

Next, for the LBP feature of each small region, a histogram representation thereof is calculated. The histogram counts the distribution of the different LBP values in the region, thereby better representing the texture features of the region.

And finally, splicing LBP histograms of all the small areas together to form an LBPH characteristic vector of the whole image. This feature vector may be used for training and recognition processes in face recognition tasks.

The LBPH identifier has the advantages of being simple to calculate, having certain robustness to illumination change and noise, and the like. The method is widely applied in the face recognition field, can be used for rapidly and accurately recognizing the face, and is suitable for a large-scale face database.

In summary, the LBPH identifier is a face recognition method based on a local binary pattern histogram, and represents a face by extracting local texture features of a face image and counting histogram representations thereof. The method has better performance and wide application in the face recognition task.

The embodiment of the invention considers that when the related technical scheme carries out teaching attendance management, the following two main problems are presented:

1) The demands of college teachers on the attendance of students are different. Basically, it is very important to check whether a teacher rolls a call or not, but because a plurality of classes are mixed to give lessons in a college, attendance can take a lot of time to give lessons, and students can delay to go early, most teachers often adopt a compromise method under the condition of no attendance system, namely, the students are checked in the lessons occasionally. Introducing an attendance system increases the management cost, time cost, to a greater or lesser extent for schools, teachers, students.

2) The defects of the traditional paper attendance table are most remarkable, and for a class with more people, the time for taking a half class is needed for completely completing one time of attendance. While the common attendance system greatly shortens the time of attendance work, students still need to actively cooperate with each class attendance, so that the teaching behavior becomes complicated in an intangible way.

Based on the above, the embodiment of the invention provides a solution, which can realize convenient and rapid classroom attendance for students in colleges and universities, solves the technical problems of low precision and low efficiency commonly existing in the current college classroom attendance system, is suitable for teaching attendance in colleges and universities, has rapid attendance process, does not occupy precious time of students and teachers, and realizes intelligent attendance.

Specifically, referring to fig. 1, fig. 1 is a schematic diagram of functional blocks of a terminal device to which the teaching attendance management device of the present invention belongs. The teaching attendance management device can be independent of a device of the terminal equipment, capable of carrying out teaching attendance management, and can be borne on the terminal equipment in a form of hardware or software. The terminal equipment can be intelligent mobile equipment with a data processing function such as a mobile phone and a tablet personal computer, and can also be fixed terminal equipment or a server with a data processing function.

In this embodiment, the terminal device of the teaching attendance management device at least includes an output module 110, a processor 120, a memory 130 and a communication module 140.

The memory 130 stores an operating system and a teaching attendance management program, and the teaching attendance management device can acquire a classroom video stream through the monitor; based on a preset student face feature library, carrying out face recognition according to the classroom video stream to obtain a face recognition result set; based on a preset attendance management algorithm, generating an attendance management result according to the face recognition result set. Information such as face recognition, attendance management result generation and the like is stored in the memory 130 through the teaching attendance management program; the output module 110 may be a display screen or the like. The communication module 140 may include a WIFI module, a mobile communication module, a bluetooth module, and the like, and communicates with an external device or a server through the communication module 140.

Wherein the teaching attendance management program in the memory 130 realizes the following steps when being executed by the processor:

acquiring a classroom video stream through a monitor;

Further, the teaching attendance management program in the memory 130 when executed by the processor also realizes the following steps:

acquiring a first face recognition result in the face recognition result set;

S10: acquiring the next face recognition result according to the time sequence;

According to the scheme, the video stream of the classroom is obtained through the monitor; based on a preset student face feature library, carrying out face recognition according to the classroom video stream to obtain a face recognition result set; based on a preset attendance management algorithm, generating an attendance management result according to the face recognition result set. Through the face feature library and the teaching attendance management algorithm, accurate teaching attendance management is realized. Therefore, intelligent attendance management is realized, and the technical problems of low teaching attendance management efficiency and low precision in the prior art are solved. Compared with the prior art, the method has the advantages of short time consumption and high efficiency.

The method embodiment of the invention is proposed based on the above-mentioned terminal equipment architecture but not limited to the above-mentioned architecture.

Referring to fig. 2, fig. 2 is a schematic flow chart of a first exemplary embodiment of the teaching attendance management method according to the present invention. The teaching attendance management method comprises the following steps:

step S101, acquiring a classroom video stream through a monitor;

the execution subject of the method of this embodiment is a teaching attendance management service system, may also be a teaching attendance management device, or may be a teaching attendance management apparatus, and this embodiment is exemplified by the teaching attendance management device, which may be integrated on an apparatus having a data processing function.

In order to solve the problems of low efficiency and low precision of the current teaching attendance management, the improvement of the precision and the efficiency of the teaching attendance management is particularly important, and in order to realize the efficient and accurate teaching attendance management, the following steps are adopted:

firstly, a special teaching attendance management device is arranged and integrated on equipment with a data processing function, and teaching attendance management can be carried out through the teaching attendance management device;

finally, through teaching attendance management device, obtain classroom video stream through the monitor, wherein, the monitor is installed in the position that needs to carry out image data acquisition, according to specific demand, can select suitable monitor type, for example installs on camera, sensor or other equipment.

Step S102, based on a preset student face feature library, carrying out face recognition according to the classroom video stream to obtain a face recognition result set;

the teaching attendance management device carries out face recognition according to a classroom video stream based on a preset student face feature library to obtain a face recognition result set, and the teaching attendance management device is realized by adopting the following steps:

firstly, the teaching attendance management device extracts face images of each frame in a video by processing a classroom video stream, a face detection algorithm can be used for identifying a face area in the video frame, and then the face feature vector of each extracted face image is extracted by using the same face feature extraction algorithm;

finally, the teaching attendance management device compares the extracted face feature vector with feature vectors in a preset student face feature library. The similarity between each face feature vector and the feature vector in the database can be calculated by using a face matching algorithm, the similarity of each face feature vector is judged according to a set threshold, and face recognition results with the similarity higher than the threshold are collected into a result set to obtain a face recognition result set.

Step S103, based on a preset attendance management algorithm, generating an attendance management result according to the face recognition result set.

The teaching attendance management device generates an attendance management result according to the face recognition result set based on a preset attendance management algorithm, and is realized by adopting the following steps:

firstly, the teaching attendance management device marks the attendance situation of each student based on a preset attendance management algorithm according to a face recognition result set, and records the specific states of the students, such as attendance, tardy, leave, and the like;

finally, the teaching attendance management device counts the number of students and the times of various attendance states, calculates related attendance indexes such as attendance rate, delay times and the like, and generates an attendance management result.

Referring to fig. 3, fig. 3 is a schematic diagram illustrating the contents of a teaching attendance management algorithm in the teaching attendance management method according to the present invention.

The teaching attendance management algorithm can call a student table, a curriculum schedule, an attendance table, a teacher table, a school building table and a camera equipment table to carry out attendance management, and can realize card punching, data maintenance and report generation of teaching attendance management.

The student list contains basic information of students, and the primary key is information such as student number, name and gender. The curriculum schedule comprises a curriculum code, a curriculum name, curriculum time and a curriculum place, and the curriculum in the watch refers to a curriculum to be arranged in a certain school period; the attendance table contains the main content of attendance, because each student may have multiple courses, each course also contains how many students and which specific class, the student number and course name and time in the attendance table are used as the main key together. The teacher's table is used for storing basic information of a teacher. The school building table stores all teaching building information and classroom information of schools. The camera equipment list is used for storing all camera information, the main key is a camera code, the camera code contains information of the classroom to which the camera equipment list belongs, and the camera equipment list also contains IP address information, so that the camera equipment list is convenient to be identified by an attendance checking system. The system does not adopt a third party library, all data are stored in a file in a text mode, the content format is standardized, and a unified model is particularly important. Each student has a name, but possibly a duplicate name, so the student identity is marked by adding a student number, which increases from 1. The system stores student pictures as a sample library for face recognition, the design adopts a naming mode of randomly generated feature codes, and the feature codes do not select student numbers because, for example, the codes of 11 and 111 are easy to be confused and have inconsistent lengths, so that the construction of students is finished. The system functions include: face recognition card punching, manager login, record checking, student management and report printing, wherein the following specific implementation process is as follows:

(1) A card punching function;

the method comprises the steps of recording a current face picture by calling a camera according to a face recognition function, comparing LBP characteristics with samples trained by a system, recording card punching time after successful card punching, and storing the recorded card punching time into a file.

(2) Maintaining data;

the data maintenance mainly comprises four operations, namely adding, deleting, changing and checking, wherein the system stores student samples by photographing, and when facing a lens, an input person can press an Enter key three times, so that three photos are obtained, and the three photos are designed to ensure the quantity of training sets and improve the recognition success rate. In the operation of deleting students, the system designs the operation of inputting verification codes so as to prevent misoperation of users, and after the deletion operation is determined, all object information is deleted, including student card punching records and photo samples.

(3) Attendance report;

the system sets the time of class and the time of class as the attendance standard, and the students can be considered to be on duty when the cards are punched before the time of class and after the time of class. The card is not punched before the lesson, but punched before the lesson, and is regarded as late. Card punching before class, but not after class, and is regarded as early return. If no card record exists, the card is regarded as absent. The administrator may select two options, daily and monthly. The daily report is selected to inquire the attendance status of any day, and as long as a student does not normally punch a card, the student is put into an abnormal list, finally a report is printed, and the list, the number of people and the like are listed in a control console. Unlike daily reports, the monthly report is a report reflecting summary, and has huge data volume, so that a process of independently generating CSV files by checking the monthly report is established, and an abnormal card-punching list is not created by the monthly report.

According to the scheme, the classroom video stream is obtained through the monitor; based on a preset student face feature library, carrying out face recognition according to the classroom video stream to obtain a face recognition result set; based on a preset attendance management algorithm, generating an attendance management result according to the face recognition result set. Through the face feature library and the teaching attendance management algorithm, accurate teaching attendance management is realized. Therefore, intelligent attendance management is realized, and the technical problems of low teaching attendance management efficiency and low precision in the prior art are solved. Compared with the prior art, the method has the advantages of short time consumption and high efficiency.

Referring to fig. 4, fig. 4 is a schematic flow chart of a second exemplary embodiment of the teaching attendance management method according to the present invention.

Based on the first exemplary embodiment, the step S102, based on a preset student face feature library, performing face recognition according to the classroom video stream, and obtaining the face recognition result set may include:

step S1021, S01: reading the frame number of the classroom video stream to obtain a classroom picture;

step S1022, S02: preprocessing the classroom picture to obtain a preprocessed classroom picture;

Step S1023, S03: carrying out face capturing on the preprocessed classroom picture through a preset face capturing algorithm to obtain face data to be recognized;

step S1024, S04: based on a preset prediction function, carrying out face recognition on the face data to be recognized according to the face feature library of the student by a preset face recognition device to obtain a face recognition result;

step S1025: repeating the steps S01-S04 until the preset timing task is finished, and obtaining a plurality of face recognition results;

step S1026: and encoding the face recognition results based on a preset time sequence encoding algorithm, and establishing the face recognition result set.

Specifically, the teaching attendance management device carries out face recognition according to a classroom video stream based on a preset student face feature library to obtain a face recognition result set, and the teaching attendance management device is realized by adopting the following steps:

first, S01: the teaching attendance management device reads the real-time frame number of the classroom video stream to obtain classroom pictures, wherein the picture formats can be jpg, png and the like;

next, S02: the teaching attendance management device preprocesses the classroom pictures to obtain preprocessed classroom pictures, wherein the preprocessing can comprise the steps of enhancing, denoising, cutting, normalizing and the like of the pictures;

Then, S03: the teaching attendance management device uses a face capturing algorithm to detect a face region of a classroom picture, finds the face region in the face region, cuts an original picture according to the position information of the face region, only retains the region related to the face, extracts characteristic information in the face region by using the face capturing algorithm, and acquires face data to be identified;

then, S04: the teaching attendance management device uses a face recognition device to extract face feature vectors from face data to be recognized, compares the feature vectors of the faces to be recognized with each face feature vector in a face feature library of students, calculates similarity or distance between the face feature vectors and each face feature vector, compares the similarity or distance with a set threshold according to the setting of a prediction function, determines whether the faces to be recognized are considered to be matched with a certain face in the face library of the students, and outputs face recognition results, possibly matched student information or unknown identities according to the comparison results;

then, the teaching attendance management device repeats the steps S01-S04 until the preset timing task is finished, and a plurality of face recognition results are obtained;

finally, the teaching attendance management device applies a time sequence coding algorithm to each face recognition result, converts the face recognition result into a corresponding coding form, records the coded result, and stores the coded face recognition result to obtain a face recognition result set.

Referring to fig. 5, fig. 5 is a schematic flow chart of constructing a face feature library in the teaching attendance management method according to the present invention.

Firstly, obtaining face photos of students;

then, creating a training set according to the face photos of the students and the face images of the standard students;

then, establishing a relation between the photos in the training set and the identity information of the students, and identifying the faces;

and finally, training according to the training set, and constructing a face feature library.

The embodiment adopts the above scheme, specifically through S01: reading the frame number of the classroom video stream to obtain a classroom picture; s02: preprocessing the classroom picture to obtain a preprocessed classroom picture; s03: carrying out face capturing on the preprocessed classroom picture through a preset face capturing algorithm to obtain face data to be recognized; s04: based on a preset prediction function, carrying out face recognition on the face data to be recognized according to the face feature library of the student by a preset face recognition device to obtain a face recognition result; S01-S04 are circulated until the preset timing task is finished, and a plurality of face recognition results are obtained; and encoding the face recognition results based on a preset time sequence encoding algorithm, and establishing the face recognition result set. And carrying out face recognition according to the classroom video stream based on a preset student face feature library to obtain a face recognition result set. Therefore, a face recognition result set is obtained, and teaching attendance management is supported.

Referring to fig. 6, fig. 6 is a schematic flow chart of a third exemplary embodiment of the teaching attendance management method according to the present invention.

Based on the first exemplary embodiment, step S103, based on a preset attendance management algorithm, generating an attendance management result according to the face recognition result set may include:

step S1031, generating a classroom management result according to the face recognition result set based on a classroom management algorithm in the attendance management algorithm;

step S1032, based on a fatigue degree detection algorithm in the attendance management algorithm, generating a student fatigue degree detection result set according to the face recognition result set, wherein the fatigue degree detection algorithm is constructed based on a gray level detection technology;

and step S1033, obtaining the attendance management result according to the classroom management result and the student fatigue detection result set.

Specifically, the teaching attendance management device generates an attendance management result according to a face recognition result set based on a preset attendance management algorithm, and the teaching attendance management device is realized by adopting the following steps:

firstly, the teaching attendance management device generates a class management result based on a class management algorithm in the attendance management algorithm according to a face recognition result set, wherein the class management result reflects attendance conditions of each student, and can provide real-time class attendance information and statistical data for teachers or management staff so as to facilitate management and analysis of attendance conditions of the students;

Then, the teaching attendance management device generates a student fatigue detection result set based on a fatigue detection algorithm in an attendance management algorithm according to a face recognition result set, wherein the fatigue detection algorithm is constructed based on a gray level detection technology, the student fatigue detection result possibly comprises student information, a fatigue score, a warning grade and the like, and the warning grade possibly comprises normal, mild fatigue, severe fatigue and the like;

finally, the teaching attendance management device obtains an attendance management result according to the classroom management result and the student fatigue detection result set.

According to the scheme, specifically, a class management result is generated according to the face recognition result set based on a class management algorithm in the attendance management algorithm; generating a student fatigue detection result set according to the face recognition result set based on a fatigue detection algorithm in the attendance management algorithm, wherein the fatigue detection algorithm is constructed based on a gray level detection technology; and obtaining the attendance management result according to the classroom management result and the student fatigue detection result set. And generating an attendance management result according to the face recognition result set by a preset attendance management algorithm. Therefore, intelligent attendance management is realized, and the technical problems of low teaching attendance management efficiency and low precision in the prior art are solved. Compared with the prior art, the method has the advantages of short time consumption and high efficiency.

Referring to fig. 7, fig. 7 is a flowchart illustrating a fourth exemplary embodiment of the teaching attendance management method according to the present invention.

Based on the above third exemplary embodiment, the step S1031, based on the class management algorithm in the attendance management algorithm, generating the class management result according to the face recognition result set may include:

step S10311, classifying the face recognition results in the face recognition result set based on a preset time sequence classification algorithm to obtain an intra-class face recognition result set, a pre-class face recognition result and a post-class face recognition result;

step S10312, based on the check-in management algorithm in the classroom management algorithm, generating a check-in result according to the pre-class face recognition result;

step S10313, based on the attendance quality detection algorithm in the classroom management algorithm, generating an attendance quality detection result set according to the intra-class face recognition result set;

step S10314, based on the sign-in management algorithm in the classroom management algorithm, generating a sign-in result according to the post-class face recognition result;

step S10315, obtaining the classroom management result according to the check-in result, the attendance quality detection result set and the check-out result.

Specifically, the teaching attendance management device generates a class management result according to the face recognition result set based on a class management algorithm in the attendance management algorithm, and the teaching attendance management device is realized by adopting the following steps:

firstly, the teaching attendance management device classifies face recognition results in a face recognition result set based on a preset time sequence classification algorithm according to timestamp information in the face recognition results to obtain an intra-class face recognition result set, a pre-class face recognition result and a post-class face recognition result;

secondly, the teaching attendance management device generates a check-in result according to a pre-class face recognition result based on a check-in management algorithm in a classroom management algorithm, marks the student as a check-in state if the student appears in the pre-class face recognition result, records related information (such as a time stamp, check-in times and the like, marks the student as an absent state if the student does not appear in the pre-class face recognition result, and records the related information;

then, the teaching attendance management device evaluates and calculates the attendance condition of the student according to the preset attendance quality detection rule and algorithm, and the consideration factors can include but are not limited to: indexes such as attendance times, attendance duration, attendance rate and the like, and possible abnormal conditions (such as repeated card punching, counterfeit others and the like) to obtain an attendance quality detection result set;

Then, the teaching attendance management device judges whether students sign in the post-class face recognition result according to a preset sign-in rule and algorithm, if the students appear in the post-class face recognition result, the students are marked as sign-in states, relevant information (such as time stamps, sign-in times and the like) is recorded, and if the students do not appear in the post-class face recognition result, the students are marked as absent states, and the relevant information is recorded;

finally, the teaching attendance management device obtains a classroom management result according to the attendance result, the attendance quality detection result set and the attendance result.

Referring to fig. 8, fig. 8 is a functional block diagram of an attendance system in the teaching attendance management method according to the present invention.

The teaching attendance system comprises a student information acquisition module, an identification module, an attendance module and a setting module.

The student information acquisition module is used for realizing student image acquisition, and the module can acquire parameters such as environment establishment image acquisition, camera installation, image storage position, image naming rules and the like, network installation, information transmission, classification and the like.

The method comprises the steps of carrying out face identification and face attendance recognition through an identification module, wherein the module is used for identifying face information, training the identified face information into a face feature library, namely a face recognition algorithm, calculating all 68 feature points of a face, carrying out parameter optimization on specific identity loss of local features, namely unrecognized features, inputting images in an image database and target images to be recognized into an optimized cross-mode face recognition model, calculating feature similarity between the target images to be recognized and the images in the image database, outputting the front N images with highest feature similarity with the target images to be recognized in the image database, and taking the images with highest feature similarity as face recognition results in classrooms at specific positions, wherein the images are not one, but a plurality of, and the expected targets are 20 students. The user observes the face image to be identified, inputs the student number corresponding to the face according to the actual situation, clicks the save button, and then the system identifies the face image according to the number to form the corresponding relation between the face image and the student number. Firstly, a haarcascade_front face_default.xml file is put into a corresponding catalog, which is a face position detection file trained by a Haar classifier, belongs to a file of OpenCV self, then frames of a video stream are intercepted by a CSI camera, three frames of the video stream are intercepted by each person as training samples in consideration of the problem of SD space size, and a naming rule is needed for distinguishing different face photos, wherein the adopted scheme is 6-bit feature codes and 8-bit random numbers. A recognizer is trained, and characteristic parameters of the face are recorded in the recognizer. In the card punching function, a camera is firstly obtained, the frame number is read in a video stream, the frame number is then converted into a gray image by utilizing a white cycle, the number of faces in a picture is captured by a Haar classifier, once the faces are recognized, a predictive function in a just-trained LBPH (location based potential of hydrogen) recognizer is immediately called to evaluate the faces, and when the number is larger than a set threshold value, a feature code white corresponding to the highest matching degree parameter is returned, and the corresponding face name can be returned due to the one-to-one correspondence of the feature code and the ID.

And the attendance module is used for realizing class attendance and student identification, and the student identification is based on the face feature database, and the identification system judges whether the identified faces in the face feature database belong to the same person. After successful recognition, the identification information (academic number) of the corresponding face in the face feature library is obtained. The classroom attendance recognition is provided according to a course information module, a classroom information module and shooting of a classroom, after students are recognized, an attendance system identifies the 'arrived' of the students, the students which are not recognized do not take any record, if the students leave the classroom after coming from the back for 20 minutes, the system can automatically restore the 'arrived' to zero, the time for attendance reaches 40 minutes according to the arrangement mode of the previous system and a camera, and the face recognition process can be called for a plurality of times: the new people are continuously identified, and the identity identification process in the attendance checking flow is continuously repeated until the lesson is finished.

The setting module realizes the entrance and exit of students, the attendance time management, the class time setting and the curriculum schedule setting, comprises the filing of the information relationship between the human face and each student, and constructing an image database, wherein the image database comprises student setting, attendance list management, classroom setting and course setting, so that the archive information of a certain student appears as long as the face of the student appears.

According to the scheme, the face recognition results in the face recognition result set are classified based on a preset time sequence classification algorithm, so that an intra-class face recognition result set, a pre-class face recognition result and a post-class face recognition result are obtained;

based on a check-in management algorithm in the classroom management algorithm, generating a check-in result according to the pre-class face recognition result; based on a attendance quality detection algorithm in the classroom management algorithm, generating an attendance quality detection result set according to the intra-class face recognition result set; based on a sign-in management algorithm in the classroom management algorithm, generating a sign-in result according to a post-class face recognition result; and obtaining the classroom management result according to the check-in result, the attendance quality detection result set and the check-out result. And generating a classroom management result according to the face recognition result set based on the classroom management algorithm. Therefore, classroom management results are obtained, and support is provided for obtaining attendance management results.

Referring to fig. 9, fig. 9 is a flowchart illustrating a fifth exemplary embodiment of a teaching attendance management method according to the present invention.

Based on the fourth exemplary embodiment, the step S10313, based on the attendance quality detection algorithm in the class management algorithm, generating the attendance quality detection result set according to the intra-class face recognition result set may include:

Step S103131, obtaining a first intra-class face recognition result in the intra-class face recognition result set;

step S103132, S05: judging whether the intra-class face recognition result is consistent with the last face recognition result;

step S103133, S06: if the in-class face recognition result is inconsistent with the last face recognition result, generating a corresponding attendance quality detection result according to the in-class face recognition result based on the attendance quality detection algorithm;

step S103134, S07: acquiring the next intra-class face recognition result according to the time sequence;

and step S103135, repeating the steps S05-S07 until the intra-class face recognition result set is traversed, and establishing the attendance quality detection result set.

Specifically, the teaching attendance management device generates a attendance quality detection result set according to the in-class face recognition result set based on the attendance quality detection algorithm in the classroom management algorithm, and the teaching attendance management device is realized by adopting the following steps:

firstly, the teaching attendance management device acquires a first intra-class face recognition result in the intra-class face recognition result set;

next, S05: the teaching attendance management device compares the intra-class face recognition result with key information of the previous face recognition result, such as the recognized face features, similarity and the like, and returns consistency if the key information is the same or the similarity is within an acceptable range; otherwise, returning inconsistent;

Then, S06: if the in-class face recognition result is inconsistent with the last face recognition result, the teaching attendance management device analyzes and judges the face recognition result based on the attendance quality detection algorithm, such as judging and identifying accuracy, facial feature quality and the like, and generates corresponding attendance quality detection results including attendance quality scoring, attendance statistics and other information, wherein if the in-class face recognition result is consistent with the last face recognition result, the next in-class face recognition result is obtained, and step S05 is executed: judging whether the intra-class face recognition result is consistent with the last face recognition result;

then, S07: the teaching attendance management device obtains the next in-class face recognition result according to the time sequence, wherein the time sequence can be determined according to a time stamp which is usually obtained from a camera or equipment and represents the shooting time of the in-class face recognition result;

finally, the teaching attendance management device repeats steps S05-S07 until the face recognition result set in the class is traversed, and a attendance quality detection result set is established according to the generated attendance quality detection result.

According to the scheme, the first intra-class face recognition result in the intra-class face recognition result set is obtained; s05: judging whether the intra-class face recognition result is consistent with the last face recognition result; s06: if the in-class face recognition result is inconsistent with the last face recognition result, generating a corresponding attendance quality detection result according to the in-class face recognition result based on the attendance quality detection algorithm; s07: acquiring the next intra-class face recognition result according to the time sequence; repeating the steps S05-S07 until the intra-class face recognition result set is traversed, and establishing the attendance quality detection result set. And generating a attendance quality detection result set according to the intra-class face recognition result set based on the attendance quality detection algorithm. Therefore, a attendance quality detection result set is obtained, and support is provided for obtaining classroom management results.

Referring to fig. 10, fig. 10 is a flowchart illustrating a sixth exemplary embodiment of a teaching attendance management method according to the present invention.

Based on the third exemplary embodiment, step S1032 is configured to generate a student fatigue detection result set according to the face recognition result set based on a fatigue detection algorithm in the attendance management algorithm, where the fatigue detection algorithm is constructed based on a gray scale detection technique and may include:

step S10321, obtaining a first set of face recognition results in the face recognition result set;

step S10322, S08: judging whether the face recognition result is matched with the previous face recognition result or not;

step S10323, S09: if the face recognition result is not matched with the previous face recognition result, analyzing the low head condition of the student in the face recognition result based on the fatigue detection algorithm to obtain a student fatigue detection result;

step S10324, S10: acquiring the next face recognition result according to the time sequence;

step S10325, repeating steps S08-S10 until the face recognition result set is traversed, and establishing the student fatigue detection result set.

Specifically, the teaching attendance management device generates a student fatigue degree detection result set according to the face recognition result set based on a fatigue degree detection algorithm in an attendance management algorithm, and the teaching attendance management device is realized by adopting the following steps:

Firstly, a teaching attendance management device acquires a first face recognition result in the face recognition result set;

next, S08: the teaching attendance management device judges whether the face recognition result is matched with the previous face recognition result, and the specific comparison mode may be different according to the difference of face recognition algorithms, but generally comprises the following steps: a. and comparing the characteristic values or the characteristic vectors of the two face recognition results. If the similarity of the feature values reaches a certain threshold, two results can be considered to be matched, b. The similarity scores of the two face recognition results are compared, some face recognition algorithms can output similarity scores, and if the score is higher than a certain threshold, the two results can be considered to be matched;

then, S09: the teaching attendance management device analyzes the low head condition of the student in the current face recognition result by using a fatigue degree detection algorithm to obtain the student fatigue degree detection result, wherein the specific fatigue degree detection algorithm may be different according to different requirements and actual scenes, but generally comprises the following steps: a. facial features of the student are extracted. The face image or the feature representation of the student can be obtained through the face recognition result, the subsequent processing is carried out by utilizing the information, and b, the fatigue detection analysis is carried out on the face features of the student, and the common method comprises the steps of analyzing based on the posture estimation, the eye state, the facial expression and the like so as to judge whether the student has the conditions of low head, fatigue and the like;

Then, S10: the teaching attendance management device obtains the next face recognition result according to the time sequence, wherein the time sequence can be determined according to a time stamp which is usually obtained from a camera or equipment and represents the shooting time of the face recognition result;

finally, the teaching attendance management device repeats the steps S08-S10 until the face recognition result set is traversed, and the student fatigue detection result set is established according to the generated student fatigue detection result.

According to the scheme, the first face recognition result in the face recognition result set is obtained; s08: judging whether the face recognition result is matched with the previous face recognition result or not; s09: if the face recognition result is not matched with the previous face recognition result, analyzing the low head condition of the student in the face recognition result based on the fatigue detection algorithm to obtain a student fatigue detection result; s10: acquiring the next face recognition result according to the time sequence; repeating the steps S08-S10 until the face recognition result set is traversed, and establishing the student fatigue degree detection result set. And generating a student fatigue detection result set according to the face recognition result set by a fatigue detection algorithm. Therefore, a student fatigue degree detection result set is obtained, and support is provided for obtaining an attendance management result.

Referring to fig. 11, fig. 11 is a flowchart of a seventh exemplary embodiment of the teaching attendance management method according to the present invention.

Based on the first exemplary embodiment, step S103 may include, based on a preset attendance management algorithm, generating an attendance management result according to the face recognition result set:

step S104, based on a preset big data processing algorithm, generating an attendance report according to the attendance management result.

Specifically, the teaching attendance management device generates an attendance report according to an attendance management result, and the teaching attendance management device is realized by the following steps:

firstly, the teaching attendance management device preprocesses attendance management results, including operations such as removing abnormal values, filling missing values and the like, so as to ensure the accuracy and the integrity of data;

finally, the teaching attendance management device performs statistics and analysis on attendance management result data according to a preset big data processing algorithm. This may involve calculating the attendance, absences, late arrival times, etc. for each student, or comparing and analyzing different classes, grades, disciplines, etc., to generate attendance reports.

According to the scheme, the attendance report is generated according to the attendance management result, specifically based on a preset big data processing algorithm. And summarizing and analyzing the obtained student fatigue detection results to generate corresponding reports or statistical data. Therefore, the attendance report is obtained, can be used for monitoring and evaluating the attendance condition of students by schools, educational institutions or managers, and provides basis for subsequent teaching management and intervention.

In addition, the invention also provides a teaching attendance management device, which comprises:

The principle of implementing the teaching attendance management in this embodiment refers to the above embodiments, and will not be described herein.

In addition, the embodiment of the invention also provides a terminal device, which comprises a memory, a processor and a teaching attendance management program stored on the memory and capable of running on the processor, wherein the teaching attendance management program realizes the steps of the teaching attendance management method when being executed by the processor.

Because the teaching attendance management program is executed by the processor, all the technical schemes of all the embodiments are adopted, at least all the beneficial effects brought by all the technical schemes of all the embodiments are provided, and the description is omitted herein.

Because the teaching attendance management program adopts all the technical schemes of all the embodiments when being executed by the processor, the teaching attendance management program at least has all the beneficial effects brought by all the technical schemes of all the embodiments and is not described in detail herein.

Compared with the prior art, the teaching attendance management scheme provided by the embodiment of the invention obtains the classroom video stream through the monitor; based on a preset student face feature library, carrying out face recognition according to the classroom video stream to obtain a face recognition result set; based on a preset attendance management algorithm, generating an attendance management result according to the face recognition result set. Thus solving the problems of low precision and low efficiency commonly existing in the classroom attendance system of colleges and universities. Based on the scheme of the invention, starting from the problems of low teaching attendance management precision and low efficiency in the real world, a teaching attendance management scheme is designed, the effectiveness of the teaching attendance management method is verified in the attendance management result, and finally the efficiency and the precision of teaching attendance management by the method are obviously improved.

Compared with the prior art, the embodiment of the invention has the following advantages:

1. high efficiency.

2. The accuracy is high.

3. The fatigue condition of students in class can be effectively detected, and students without class and midway early-retreating are identified.

4. The background database and the monitoring video management system can be linked according to teaching arrangement of different classes, so that intelligent attendance for different classes in the same teaching period is realized.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.

The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) as above, comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, a controlled terminal, or a network device, etc.) to perform the method of each embodiment of the present invention.

Claims

1. The teaching attendance management method is characterized by comprising the following steps of:

acquiring a classroom video stream through a monitor;

2. The teaching attendance management method as claimed in claim 1, wherein the step of performing face recognition according to the classroom video stream based on a preset student face feature library to obtain a face recognition result set comprises:

3. The teaching attendance management method as claimed in claim 1, wherein the step of generating an attendance management result based on the face recognition result set based on a preset attendance management algorithm comprises:

4. The teaching attendance management method as claimed in claim 3, wherein the step of generating a classroom management result according to the face recognition result set based on a classroom management algorithm in the attendance management algorithm comprises:

5. The teaching attendance management method as claimed in claim 4, wherein the step of generating a attendance quality detection result set based on the attendance quality detection algorithm in the class management algorithm according to the intra-class face recognition result set comprises:

6. The teaching attendance management method as claimed in claim 3, wherein the step of generating a student fatigue detection result set based on the fatigue detection algorithm in the attendance management algorithm according to the face recognition result set comprises:

acquiring a first face recognition result in the face recognition result set;

S10: acquiring the next face recognition result according to the time sequence;

7. The teaching attendance management method as claimed in any one of claims 1-6, wherein the step of generating an attendance management result based on the face recognition result set based on a preset attendance management algorithm further comprises:

8. The utility model provides a teaching attendance management device which characterized in that, teaching attendance management device includes:

9. A terminal device comprising a memory, a processor and a teaching attendance management program stored on the memory and operable on the processor, the teaching attendance management program when executed by the processor implementing the steps of the teaching attendance management method as claimed in any of claims 1 to 7.

10. A computer readable storage medium, wherein a teaching attendance management program is stored on the computer readable storage medium, and when executed by a processor, the teaching attendance management program implements the steps of the teaching attendance management method as claimed in any one of claims 1 to 7.