CN117036117B - Classroom state assessment method based on artificial intelligence - Google Patents
Classroom state assessment method based on artificial intelligence Download PDFInfo
- Publication number
- CN117036117B CN117036117B CN202310698478.5A CN202310698478A CN117036117B CN 117036117 B CN117036117 B CN 117036117B CN 202310698478 A CN202310698478 A CN 202310698478A CN 117036117 B CN117036117 B CN 117036117B
- Authority
- CN
- China
- Prior art keywords
- student
- node
- teaching
- head
- students
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 32
- 230000008859 change Effects 0.000 claims abstract description 71
- 238000011156 evaluation Methods 0.000 claims description 35
- 210000000988 bone and bone Anatomy 0.000 claims description 18
- 230000033764 rhythmic process Effects 0.000 claims description 7
- 230000006870 function Effects 0.000 claims description 6
- 230000008447 perception Effects 0.000 claims description 5
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000006399 behavior Effects 0.000 abstract description 10
- 238000012545 processing Methods 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 10
- 230000009471 action Effects 0.000 description 5
- 238000013527 convolutional neural network Methods 0.000 description 5
- 238000012549 training Methods 0.000 description 4
- 241001310793 Podium Species 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000013480 data collection Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/34—Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Strategic Management (AREA)
- Computational Linguistics (AREA)
- Tourism & Hospitality (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Educational Technology (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Educational Administration (AREA)
- Medical Informatics (AREA)
- Databases & Information Systems (AREA)
- Marketing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Business, Economics & Management (AREA)
- Primary Health Care (AREA)
- Economics (AREA)
- Human Computer Interaction (AREA)
- Human Resources & Organizations (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to the technical field of data processing for supervision, in particular to an artificial intelligence-based classroom state assessment method. The classroom state assessment method based on artificial intelligence comprises the following steps: acquiring voice data of a teacher through recording equipment; acquiring a start signal entering the teaching node and an end signal leaving the teaching node, which are sent by a teacher, according to the voice data; obtaining a picture of the position of a student in a current teaching node through a depth camera; acquiring the head posture change of each student in the current teaching node according to the photo of the student position; and judging whether the students listen to lessons in the current teaching node according to the head posture change of each student. According to the application, the time period guided by the teacher is identified as the teaching node, and whether the behaviors of the students are sufficiently guided by the teacher is detected in the teaching nodes, so that whether the students are listening to the lesson is detected, and the accuracy of classroom state assessment can be improved.
Description
Technical Field
The application relates to the technical field of data processing for supervision, in particular to an artificial intelligence-based classroom state assessment method.
Background
In CN110503000a, a method for acquiring a class state by detecting a head-up rate is proposed, and the method mainly focuses on detecting whether a student has a whole head-up or not.
But the normal course of classroom teaching includes several teaching processes of teacher lecture, student taking notes, student asking questions, teacher asking questions and teacher answering, etc. Meanwhile, in the whole classroom learning process, due to different personal habits of students, some students like to raise the head to watch teacher courseware to listen to lessons, the head raising rate of the students in the whole classroom can be higher, some students like to listen to lessons against books, and the head raising rate of the students in the whole classroom can be lower. Therefore, the method for directly evaluating whether the class state is good or not is not accurate enough by acquiring the head-up rate of students in the whole class.
The accuracy of the state evaluation of the class in the prior art needs to be improved
Disclosure of Invention
In order to solve the technical problems or at least partially solve the technical problems, the application provides an artificial intelligence-based classroom state assessment method, which can more accurately assess and reflect the classroom state.
In a first aspect, the present application provides an artificial intelligence based class state evaluation method, including:
acquiring voice data of a teacher through recording equipment;
acquiring a start signal entering the teaching node and an end signal leaving the teaching node, which are sent by a teacher, according to the voice data;
obtaining a picture of the position of a student in a current teaching node through a depth camera;
Acquiring the head posture change of each student in the current teaching node according to the photo of the student position;
and judging whether the students listen to lessons in the current teaching node according to the head posture change of each student.
Optionally, the classroom evaluation method based on artificial intelligence further includes:
detecting the teaching state of the teaching node after leaving the current teaching node;
the detecting the teaching state of the teaching node after leaving the current teaching node comprises:
and in a first preset time after the teacher sends out an end signal leaving the current teaching node, acquiring the head posture change of the student in the first preset time through the photo of the position of the student, acquiring the proportion of the student turning left or right in the first preset time according to the head posture change of the student in the first preset time, and judging that the teaching rhythm of the teaching node is too fast when the proportion of the student is larger than a first preset threshold value.
Optionally, the classroom state assessment method based on artificial intelligence further includes:
Acquiring the head posture change of the student in a second preset time before the teacher sends out an end signal leaving the current teaching node;
According to the head posture change of the student in the second preset time, the student rate of left turning or right turning of the student in the second preset time is obtained, and when the student rate of the student in the second preset time is larger than a second preset threshold value, the teaching rhythm of the teaching node is judged to be too slow.
Optionally, the classroom state assessment method based on artificial intelligence further includes:
When the total number of teaching nodes of the students which do not attend the class exceeds a preset threshold value of not attending the class, judging that the students do not attend the class in the current class;
the method comprises the steps of obtaining a class listening state evaluation value of a current class through a preset class listening state evaluation function, wherein the preset class listening state evaluation function is as follows:
wherein P is an assessment value of class listening state, N is the total number of students, and M is the total number of students who do not listen to class.
Optionally, the classroom evaluation method based on artificial intelligence further includes:
the current classroom state evaluation value is evaluated through a preset classroom state evaluation formula, wherein the classroom state evaluation formula is as follows:
Wherein, TSE is classroom state evaluation value, node all is teaching Node total number, node good is excellent teaching Node total number, node fast is teaching Node total number with too fast teaching speed, node slow is teaching Node total number with too slow teaching speed, and Node optimal is teaching Node finger derivative;
In the middle of Sigma is the tolerance constant.
Optionally, the obtaining, by the depth camera, a photograph of the student location includes:
The depth camera comprises a first depth camera arranged in front of the classroom and a second depth camera arranged behind the classroom;
The first depth camera is used for acquiring a front photo of the student position, the second depth camera is used for acquiring a rear photo of the student position, and the first depth camera and the second depth camera are used for shooting the student position at the same time to obtain a first depth image and a second depth image respectively.
Optionally, obtaining the head posture change of each student in the current teaching node according to the photo of the student position includes:
Acquiring human skeleton key points of the left eye, the right eye, the nose and the neck of each student through AlphaPose algorithm according to the first depth image;
acquiring human skeleton key points of the neck, the left shoulder, the right shoulder and the back of each student through AlphaPose algorithm according to the second depth image;
The neck is used as a joint connecting point to splice the human skeleton key points obtained by the first depth image and the human skeleton key points obtained by the second depth image to obtain skeleton forms of each student;
Converting other bone key points into coordinate values taking the neck bone key points as coordinate origins by taking the neck bone key points in the bone morphology as centers;
Collecting coordinate values of all converted skeleton key points of each student to form a gesture description set of each student, and inputting the gesture description set of each student into a trained gesture perception model to obtain the head gesture of each student;
And shooting the pictures of the positions of the students in the current teaching node for multiple times by the first depth camera and the second depth camera according to the preset interval time, so that the head posture change of each student can be obtained.
Optionally, determining whether the student listens to the lesson in the current teaching node according to the head posture change of the student includes:
Respectively acquiring a preset number of head gestures for each student, and sequencing according to the occurrence time to serve as a head gesture change sequence, wherein the head gestures in the head gesture change sequence are head gestures of each student which continuously occur in a current teaching node and the number of the continuous occurrence times is forefront;
Acquiring the head posture change sequence with the largest occurrence number in the whole students according to the head posture change sequence of each student as a head change sequence standard;
And comparing the head change sequence of each student with the head change sequence standard, and judging that the current teaching node of the student does not listen to lessons when the head posture change sequence of the student is not in accordance with the head posture change sequence standard.
Optionally, the preset number is 2.
In a second aspect, the present application provides a computer readable storage medium comprising the artificial intelligence based classroom state assessment method of any of the first aspects.
Compared with the prior art, the technical scheme provided by the application has the following advantages:
In the teaching process of a classroom, when a teacher encounters important and difficult knowledge points, the teacher can guide students to make certain behaviors through specific actions and voices, for example, when the teacher teaches important knowledge, the teacher can give attention to a blackboard so as to guide the concentration of the attention of the students, and thus the rhythm of the whole teaching is mastered. In the teaching process, the students make behaviors, and the behaviors are guided by teachers, so that the consistency of the behaviors is high. In other processes of the classroom, which are not guided by teachers, even if students get low or turn around, the students are affected by personal habits, it cannot be fully stated whether they are not listening to the lessons.
Therefore, the teaching node is used for identifying the teaching time periods guided by the teachers and detecting whether the behaviors of the students are sufficiently guided by the teachers or not in the teaching nodes, so that whether the students are in class or not is detected, and the accuracy of classroom state assessment can be improved.
Drawings
Fig. 1 is a schematic diagram of an application scenario of an artificial intelligence-based classroom state assessment method according to an embodiment of the present application;
fig. 2 is a flowchart of an artificial intelligence-based classroom state assessment method according to an embodiment of the present application.
Detailed Description
The technical scheme of the application will be described below with reference to the accompanying drawings.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application, but the present application may be practiced otherwise than as described herein; it will be apparent that the embodiments in the specification are only some, but not all, embodiments of the application. It should be noted that, without conflict, the embodiments of the present application and features in the embodiments may be combined with each other.
Fig. 1 is a schematic diagram of an application scenario of an artificial intelligence-based classroom state assessment method according to an embodiment of the present application.
As shown in fig. 1, the classroom state assessment method based on artificial intelligence provided by the application is generally applied to classroom teaching in a middle and primary school education stage, and is used for assessing the classroom teaching state of teachers, so as to provide fine index references for the teachers and schools, and therefore, the teaching mode is adjusted according to the index to achieve a better teaching effect.
As shown in fig. 1, the classroom state assessment method based on artificial intelligence provided by the application uses a first depth camera, a second depth camera and a recording device, wherein the recording device is used for collecting voice information of a teacher, and identifying whether a teaching node starts or not through a contracted starting instruction realized by the teacher, for example, the starting instruction is 'the attention of students', and when the configured voice identification module identifies that the teacher has sent a starting signal entering the teaching node through the recording device, the teaching node is judged to start. And shooting the positions of the students at intervals of a preset interval time simultaneously by the first depth camera and the second depth camera to obtain a plurality of first depth images and second depth images, extracting skeleton forms of each student by the first depth images and the second depth images, and obtaining head posture changes of each student after the start of teaching nodes by a trained posture perception model.
For example, the end instruction is "we proceed with the explanation of the next knowledge point". And when the voice recognition module recognizes the sentence through the recording equipment, judging that the teaching node is ended. And then detecting whether the head posture change of each student in the teaching node is inconsistent with the head posture change of all students, so that students who can listen to class and cannot listen to class can be identified.
In this embodiment, the voice recognition module may be a computer installed on a podium, and processes voice information by using a conventional voice recognition algorithm or a commercial voice recognition structure, thereby recognizing a contracted start instruction and end instruction.
In other embodiments, a similarity comparison may be made between the voice information identified by the voice recognition module and the contracted start instruction and end instruction, and by setting a similarity threshold, the teacher does not have to die from the start instruction and end instruction of the teaching node.
According to the classroom state assessment method based on artificial intelligence, provided by the embodiment of the application, the teaching time periods guided by teachers can be identified and obtained as teaching nodes through the contracted starting instruction and ending instruction. And in these teaching nodes it is detected whether the student's behaviour is sufficiently guided by the teacher, and thus whether the student is listening to the class. The probability of inaccurate recognition of the student class listening state caused by different class habits of students is reduced, so that the accuracy of class state evaluation can be improved.
In this embodiment, the classroom state assessment method based on artificial intelligence includes:
S201: and acquiring voice data of a teacher through the recording equipment.
Specifically, the recording device includes a microphone, and the microphone is disposed on a platform to obtain voice information of a teacher.
S202: and acquiring a start signal which is sent by a teacher and enters the teaching node and an end signal which is sent by the teacher and leaves the teaching node according to the voice data.
Specifically, in this embodiment, the voice information collected by the recording device is processed by the computer provided in the podium. The start signal and the end signal can be freely set by the teacher. In this embodiment, the computer processes the voice information of the classroom through the voice recognition interface to recognize whether the teacher has issued the start signal and the end signal.
S203: and obtaining a picture of the student position in the current teaching node through the depth camera.
Specifically, the obtaining, by the depth camera, a photograph of the student's location includes:
The depth camera comprises a first depth camera arranged in front of the classroom and a second depth camera arranged behind the classroom;
The first depth camera is used for acquiring a front photo of the student position, the second depth camera is used for acquiring a rear photo of the student position, and the first depth camera and the second depth camera are used for shooting the student position at the same time to obtain a first depth image and a second depth image respectively.
In this embodiment, the front photograph refers to a photograph that can include the face of the student, and the rear photograph refers to a photograph that can include the back of the student.
In a class node, the depth camera photographs the student position every preset interval time to obtain a plurality of first depth images and second depth images.
The preset interval is 5 seconds, and in other embodiments, the preset interval may be adjusted by a worker.
S204: and obtaining the head posture change of each student in the current teaching node according to the photo of the student position.
Specifically, obtaining the head posture change of each student in the current teaching node according to the photo of the student position includes:
And respectively acquiring human skeleton key points of the left eye, the right eye, the nose and the neck of each student through AlphaPose algorithm according to the first depth image.
And respectively acquiring human skeleton key points of the neck, the left shoulder, the right shoulder and the back of each student through AlphaPose algorithm according to the second depth image.
Specifically, the ALphPose algorithm utilized in this embodiment is a prior art.
AlphaPose is to use the deep convolutional neural network to perform feature extraction and key point regression on the human body image. The algorithm uses two main modules to accomplish pose estimation: one is a target detection module for human body detection, which is used for identifying the position of the human body in the image; the other is a gesture estimation module for key point regression, which is used for deducing the key point position of the human body.
And splicing the human skeleton key points obtained by the first depth image and the human skeleton key points obtained by the second depth image by taking the neck as a joint connecting point to obtain the skeleton form of each student.
And converting other bone key points into coordinate values taking the neck bone key points as coordinate origins by taking the neck bone key points in the bone morphology as centers.
Specifically, the coordinates of other bone key points in the space point cloud coordinate system are converted into relative coordinates with the neck key points as the original points, and the converted coordinates can be obtained by directly using the difference between the coordinates in the original space point cloud coordinate system and the neck key points.
Collecting coordinate values of all converted skeleton key points of each student to form a gesture description set of each student, and inputting the gesture description set of each student into a trained gesture perception model to obtain the head gesture of each student.
The set of gesture descriptions PS in this embodiment is:
PS=[(k0,x0,y0,z0);(k1,x1,y1,z1);(k2,x2,y2,z2);(k3,x3,y3,z3);…;(k6,x6,y6,z6)]
Wherein the array represented by k 0-k6 represents the key point marks of the human bones of the neck, the left eye, the right eye, the nose, the left shoulder, the right shoulder and the back and the transformed point cloud coordinates thereof.
The head posture change of each student in the teaching node can be obtained by shooting pictures of the positions of the students in the current teaching node for multiple times through the first depth camera and the second depth camera according to preset interval time and inputting each picture into a trained posture sensing model.
Specifically, the gesture perception model is obtained through training the following steps:
Data collection and labeling: a human skeletal keypoint dataset comprising head poses of head up, head down, head left turn, head right turn is collected. In each dataset sample, the coordinates of each keypoint are included and associated with a corresponding head pose.
Data preprocessing: and normalizing the key point coordinates, namely converting the point cloud coordinates of the bone key points into point cloud coordinates taking the neck bone key points as the centers.
Model selection: in this embodiment, convolutional Neural Network (CNN) is used for training.
Model training: the model is trained using the annotated dataset. And taking the key point data as input, and taking the corresponding head gesture as a target to perform supervised learning. The weights and parameters of the model are optimized through iteration, so that the model can accurately predict the action category of the given gesture description set.
Model evaluation and tuning: and evaluating the performance of the CNN model by using a test data set independent of the training set, and adjusting the super parameters of the CNN model according to the obtained result.
According to the embodiment of the application, the first depth camera and the second depth camera which are respectively arranged in front of a classroom and behind a teacher are used for extracting the bone morphology of the student, so that the situation that the bone morphology below the student cannot be extracted due to shielding of a desk can be effectively avoided, and the student can rotate the trunk generally, so that the morphology information can be fully captured through the first depth cameras respectively arranged in front of the student and behind the student, and the accuracy of recognizing the head gesture of the student is improved.
S205: and judging whether the students listen to lessons in the current teaching node according to the head posture change of each student.
Specifically, judging whether the student listens to the lesson in the current teaching node according to the head posture change of the student comprises:
And respectively acquiring a preset number of head gestures for each student, and sequencing according to the occurrence time to serve as a head gesture change sequence, wherein the head gestures in the head gesture change sequence are head gestures of each student which continuously occur in the current teaching node and the number of the continuous occurrence times is the forefront.
And acquiring the head posture change sequence with the largest occurrence number in the whole students according to the head posture change sequence of each student as a head change sequence standard.
And comparing the head change sequence of each student with the head change sequence standard, and judging that the current teaching node of the student does not listen to lessons when the head posture change sequence of the student is not in accordance with the head posture change sequence standard.
For example, in this embodiment, the posture sensing model can identify that the head posture of the student is a: head raising, B: low head, C: left turn head or D: and (5) turning the head right. The preset number is 2.
The teaching node is a teacher question asking link, so after entering the teaching node, a general student can first lower his head to do questions and then raise his head to see the questions to a teacher.
The pose of each student head within a teaching node is changed to:
Student 1: [ A, B, B, B, A, A, A, B, C ];
Wherein, B appears continuously at most 3 times, A appears continuously at most 3 times, and the head change sequence of the student 1 is [ B, A ] according to the time sequence of appearance.
Student 2: [ C, D, A, A, B, B, A, A, A ];
Wherein, B appears continuously at most 2 times, A appears continuously at most 3 times, and the head gesture change sequence of the student 2 is [ B, A ] according to the appearance time sequence of the head gestures appearing continuously.
Student 3: [ A, A, A, B, B, B, A, A ];
Wherein, B appears continuously at most 3 times, A appears continuously at most 4 times, and the head gesture change sequence of the student 3 is [ A, B ] according to the appearance time sequence of the head gestures appearing continuously.
Next, the head change order [ B, a ] appears 2 times, and the head change order [ a, B ] appears 1 time, and then the head change order with the largest appearance number among all students is selected as the head change order standard.
Finally, the head change sequence of each student is compared with the head change sequence standard, so that the students 3 can learn that the students 1 and 2 do not listen to the lessons.
The embodiment of the application adopts the head gestures with the largest number of continuous occurrence times for mutual comparison, can ensure that the main behavior change of the student is extracted, and improves the accuracy of the comparison of the head gestures of the student.
In other embodiments, the preset number may be adjusted by a worker, the larger the number, the lower the tolerance to inconsistent changes in student behavior.
Meanwhile, in other embodiments, a plurality of head change sequences can be acquired from all students as head change sequence standards for comparison, so that tolerance to inconsistent behaviors of the students can be improved.
In this embodiment, the classroom evaluation method based on artificial intelligence further includes:
detecting the teaching state of the teaching node after leaving the current teaching node;
the detecting the teaching state of the teaching node after leaving the current teaching node comprises:
and in a first preset time after the teacher sends out an end signal leaving the current teaching node, acquiring the head posture change of the student in the first preset time through the photo of the position of the student, acquiring the proportion of the student turning left or right in the first preset time according to the head posture change of the student in the first preset time, and judging that the teaching rhythm of the teaching node is too fast when the proportion of the student is larger than a first preset threshold value.
In this embodiment, the first preset time is 1 minute, and the first preset threshold is 40%. If the left turning head or the right turning head of the student with the first preset threshold appears within 1 minute after the current teaching node is finished, the fact that most students need to be discussed with students to complete notes or fully understand the students can be indicated, and the teaching speed of the teaching node can be indicated to be too high.
In this embodiment, the classroom state assessment method based on artificial intelligence further includes:
And acquiring the head posture change of the student in a second preset time before the teacher sends out the ending signal leaving the current teaching node.
According to the head posture change of the student in the second preset time, the student rate of left turning or right turning of the student in the second preset time is obtained, and when the student rate of the student in the second preset time is larger than a second preset threshold value, the teaching rhythm of the teaching node is judged to be too slow.
In this embodiment, the second preset time is 1 minute, and the second preset threshold is 40%. If the left turning head or the right turning head of the student with the second preset threshold appears within 1 minute before the current teaching node is finished, the teaching node can be used for indicating that most students are distracted and the teaching speed of the teaching node is too slow.
In this embodiment, the classroom state assessment method based on artificial intelligence further includes:
And when the total number of teaching nodes of the student which do not attend the class exceeds a preset threshold value of not attending the class, judging that the student does not attend the class in the current class.
The specific threshold value of not listening to lessons can be set by a teacher, and is generally set to be 30% -60% of the total number of the current teaching nodes.
The method comprises the steps of obtaining a class listening state evaluation value of a current class through a preset class listening state evaluation function, wherein the preset class listening state evaluation function is as follows:
wherein P is an assessment value of class listening state, N is the total number of students, and M is the total number of students who do not listen to class.
In this embodiment, the classroom evaluation method based on artificial intelligence further includes:
the current classroom state evaluation value is evaluated through a preset classroom state evaluation formula, wherein the classroom state evaluation formula is as follows:
The TSE is a class state evaluation value, node all is a total number of teaching nodes, node good is a total number of excellent teaching nodes, node fast is a total number of teaching nodes with too high teaching speed, node slow is a total number of teaching nodes with too low teaching speed, and Node optimal is a derivative of teaching nodes.
In the middle ofSigma is the tolerance constant.
Specifically, the total number of teaching nodes is the sum of all teaching nodes identified in the current class, and the total number of excellent teaching nodes is the total number of teaching nodes, namely Node good=Nodeall-Nodefast-Nodeslow, the total number of teaching nodes with too high teaching speed, namely the total number of teaching nodes with too low teaching speed.
Specifically, the teaching node index derivative can be set by a school to arrange teaching node indexes of each section of courses.
Specifically, the deviation tolerance value constant is set by a worker, and when the numerical value is larger, the deviation tolerance degree of the total number of the current class teaching nodes and the teaching instruction derivative is larger.
The value range of the classroom state evaluation value is 0-P, the classroom state of the whole class is reflected to be worse when the classroom state evaluation value is closer to 0, and the classroom state of the whole class is reflected to be better when the classroom state evaluation value is closer to P.
The embodiment of the application also provides a computer readable storage medium, which comprises the classroom state evaluation method based on artificial intelligence.
It should be noted that in this document, relational terms such as "first" and "second" and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. In addition, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus 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 apparatus. 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 apparatus that comprises the element. Moreover, in the description of the embodiments of the present application, unless otherwise indicated, "/" means or, for example, a/B may mean a or B; "and/or" herein is merely an association relationship describing an association object, and means that three relationships may exist, for example, a and/or B may mean: a exists alone, A and B exist together, and B exists alone. Also, in the description of the embodiments of the present application, "plurality" means two or more than two.
The foregoing is only a specific embodiment of the application to enable those skilled in the art to understand or practice the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown and described herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (4)
1. The classroom state assessment method based on the artificial intelligence is characterized by comprising the following steps of:
acquiring voice data of a teacher through recording equipment;
Acquiring a start signal which is sent by a teacher and enters the teaching node and an end signal which is sent by the teacher and leaves the teaching node according to the voice data, so as to start acquiring the head posture change of the student of the current teaching node after the start signal is sent, and stopping acquiring the head posture change of the student of the current teaching node after the end signal is sent;
obtaining a picture of the position of a student in a current teaching node through a depth camera;
Acquiring the head posture change of each student in the current teaching node according to the photo of the student position;
judging whether the students in the current teaching node listen to lessons or not according to the head posture change of each student;
Judging whether the students listen to lessons in the current teaching node according to the head posture changes of the students comprises the following steps:
Respectively acquiring a preset number of head gestures for each student, and sequencing according to the occurrence time to serve as a head gesture change sequence, wherein the head gestures in the head gesture change sequence are head gestures of each student which continuously occur in a current teaching node and the number of the continuous occurrence times is forefront;
Acquiring the head posture change sequence with the largest occurrence number in the whole students according to the head posture change sequence of each student as a head change sequence standard;
comparing the head change sequence of each student with the head change sequence standard, and judging that the current teaching node of the student does not listen to lessons when the head posture change sequence of the student is not in accordance with the head posture change sequence standard;
the classroom evaluation method based on artificial intelligence further comprises the following steps:
detecting the teaching state of the teaching node after leaving the current teaching node;
the detecting the teaching state of the teaching node after leaving the current teaching node comprises:
acquiring the head posture change of a student in a first preset time through the photo of the position of the student in a first preset time after a teacher sends out an end signal leaving a current teaching node, acquiring the proportion of the student turning left or right in the first preset time according to the head posture change of the student in the first preset time, and judging that the teaching rhythm of the teaching node is too fast when the proportion of the student is larger than a first preset threshold value;
The classroom state assessment method based on artificial intelligence further comprises the following steps:
Acquiring the head posture change of the student in a second preset time before the teacher sends out an end signal leaving the current teaching node;
According to the head posture change of the students in the second preset time, obtaining the student rate of left turning or right turning of the students in the second preset time, and judging that the teaching rhythm of the teaching node is too slow when the student rate of the students in the second preset time is larger than a second preset threshold;
The classroom state assessment method based on artificial intelligence further comprises the following steps:
When the total number of teaching nodes of the students which do not attend the class exceeds a preset threshold value of not attending the class, judging that the students do not attend the class in the current class;
the method comprises the steps of obtaining a class listening state evaluation value of a current class through a preset class listening state evaluation function, wherein the preset class listening state evaluation function is as follows:
，
Wherein P is an assessment value of a class listening state, N is the total number of students, and M is the total number of students who do not listen to class;
the classroom evaluation method based on artificial intelligence further comprises the following steps:
the current classroom state evaluation value is evaluated through a preset classroom state evaluation formula, wherein the classroom state evaluation formula is as follows:
，
Wherein, TSE is classroom state evaluation value, node all is teaching Node total number, node good is excellent teaching Node total number, node fast is teaching Node total number with too fast teaching speed, node slow is teaching Node total number with too slow teaching speed, node optimal is teaching Node finger derivative, node good=Nodeall-Nodefast-Nodeslow;
In the middle of Sigma is the tolerance constant.
2. The artificial intelligence based classroom state assessment method of claim 1 wherein the capturing a photograph of the student's location with a depth camera includes:
The depth camera comprises a first depth camera arranged in front of the classroom and a second depth camera arranged behind the classroom;
The first depth camera is used for acquiring a front photo of the student position, the second depth camera is used for acquiring a rear photo of the student position, and the first depth camera and the second depth camera are used for shooting the student position at the same time to obtain a first depth image and a second depth image respectively.
3. The artificial intelligence based classroom state assessment method of claim 2 wherein obtaining a change in head pose of each student in a current teaching node from a photograph of the student's location comprises:
Acquiring human skeleton key points of the left eye, the right eye, the nose and the neck of each student through AlphaPose algorithm according to the first depth image;
acquiring human skeleton key points of the neck, the left shoulder, the right shoulder and the back of each student through AlphaPose algorithm according to the second depth image;
The neck is used as a joint connecting point to splice the human skeleton key points obtained by the first depth image and the human skeleton key points obtained by the second depth image to obtain skeleton forms of each student;
Converting other bone key points into coordinate values taking the neck bone key points as coordinate origins by taking the neck bone key points in the bone morphology as centers;
Collecting coordinate values of all converted skeleton key points of each student to form a gesture description set of each student, and inputting the gesture description set of each student into a trained gesture perception model to obtain the head gesture of each student;
And shooting the pictures of the positions of the students in the current teaching node for multiple times by the first depth camera and the second depth camera according to the preset interval time, so that the head posture change of each student can be obtained.
4. The artificial intelligence based classroom state assessment method of claim 1 wherein the predetermined number is 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310698478.5A CN117036117B (en) | 2023-06-14 | 2023-06-14 | Classroom state assessment method based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310698478.5A CN117036117B (en) | 2023-06-14 | 2023-06-14 | Classroom state assessment method based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117036117A CN117036117A (en) | 2023-11-10 |
| CN117036117B true CN117036117B (en) | 2024-05-24 |
Family
ID=88643622
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310698478.5A Active CN117036117B (en) | 2023-06-14 | 2023-06-14 | Classroom state assessment method based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117036117B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107609517A (en) * | 2017-09-15 | 2018-01-19 | 华中科技大学 | A kind of classroom behavior detecting system based on computer vision |
| CN110503000A (en) * | 2019-07-25 | 2019-11-26 | 杭州电子科技大学 | A kind of teaching new line rate measurement method based on face recognition technology |
| WO2020118669A1 (en) * | 2018-12-11 | 2020-06-18 | 深圳先进技术研究院 | Student concentration detection method, computer storage medium, and computer device |
| CN116228478A (en) * | 2023-02-10 | 2023-06-06 | 深圳市沃特沃德信息有限公司 | Student card-based learning quality assessment method, device, equipment and medium |
-
2023
- 2023-06-14 CN CN202310698478.5A patent/CN117036117B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107609517A (en) * | 2017-09-15 | 2018-01-19 | 华中科技大学 | A kind of classroom behavior detecting system based on computer vision |
| WO2020118669A1 (en) * | 2018-12-11 | 2020-06-18 | 深圳先进技术研究院 | Student concentration detection method, computer storage medium, and computer device |
| CN110503000A (en) * | 2019-07-25 | 2019-11-26 | 杭州电子科技大学 | A kind of teaching new line rate measurement method based on face recognition technology |
| CN116228478A (en) * | 2023-02-10 | 2023-06-06 | 深圳市沃特沃德信息有限公司 | Student card-based learning quality assessment method, device, equipment and medium |
Non-Patent Citations (1)
| Title |
|---|
| 人脸检测和表情识别在课堂教学评价中的应用研究;梁利亭;山东商业职业技术学院学报;第20卷(第02期);第100-104页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117036117A (en) | 2023-11-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110991381B (en) | Real-time classroom student status analysis and indication reminding system and method based on behavior and voice intelligent recognition | |
| CN108648757B (en) | Analysis method based on multi-dimensional classroom information | |
| CN110992741B (en) | Learning auxiliary method and system based on classroom emotion and behavior analysis | |
| CN107230174B (en) | Online interactive learning system and method based on network | |
| CN108399376A (en) | Student classroom learning interest intelligent analysis method and system | |
| CN115239527B (en) | Teaching behavior analysis system based on knowledge base teaching feature fusion and modeling | |
| CN109933650B (en) | Method and system for understanding picture title in operation | |
| CN108765229B (en) | Learning performance evaluation method based on big data and artificial intelligence and robot system | |
| CN108829842B (en) | Learning expression image method and robot system based on big data and artificial intelligence | |
| CN113065757A (en) | Method and device for evaluating on-line course teaching quality | |
| CN114021962A (en) | Teaching evaluation method, evaluation device and related equipment and storage medium | |
| CN113888757A (en) | Examination paper intelligent analysis method, examination paper intelligent analysis system and storage medium based on benchmarking evaluation | |
| CN112085392A (en) | Learning participation degree determining method and device and computer equipment | |
| WO2022009875A1 (en) | Program for determining concentration level | |
| Seneviratne et al. | Student and lecturer performance enhancement system using artificial intelligence | |
| CN108876677A (en) | Assessment on teaching effect method and robot system based on big data and artificial intelligence | |
| CN117036117B (en) | Classroom state assessment method based on artificial intelligence | |
| CN110598607B (en) | Non-contact and contact cooperative real-time emotion intelligent monitoring system | |
| KR101996247B1 (en) | Method and apparatus of diagnostic test | |
| CN115544226B (en) | Similarity recognition method based on multi-mode emotion analysis | |
| CN110991943A (en) | Teaching quality evaluation system based on cloud computing | |
| CN110826796A (en) | Score prediction method | |
| CN111667128A (en) | Teaching quality assessment method, device and system | |
| CN115689340A (en) | Classroom quality monitoring system based on colorful dynamic human face features | |
| CN115171042A (en) | Student classroom behavior identification method, device, terminal equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |