Intelligent classroom teaching management system based on Internet
By modeling students' locations and video stream analysis, obtaining action data and calculating concentration analysis coefficients, the accuracy and targeted problems of concentration monitoring in the existing smart teaching management system are solved, and the efficiency of teaching management is improved.
Patent Information
- Application Number
- CN202510408426.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-02
- Publication Date
- 2025-07-18
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing smart teaching management system cannot effectively monitor students' concentration, and lacks targetedness and accuracy, resulting in inefficiency in teaching.
By spatially modeling students' locations, real-time video stream acquisition and analysis, periodic action data are obtained, and concentration analysis coefficients are calculated, precise monitoring and management of students' concentration is achieved.
It improves the accuracy and pertinence of teaching management, ensures the accuracy and pertinence of concentration monitoring results, and improves teaching efficiency.
Smart Images
Figure CN120339002A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of intelligent teaching, relates to image processing technology, and specifically is an intelligent teaching management system based on an Internet classroom. Background Art
[0002] When the existing intelligent teaching management system manages students, the following specific defects exist:
[0003] 1. When the existing intelligent teaching management system manages students, when it is unable to monitor the concentration of students through monitoring devices, it can only monitor the large-scale body movements of students through a single monitoring image, and the monitoring results are difficult to effectively reflect the learning status of students, thus easily leading to low teaching efficiency;
[0004] 2. When the existing intelligent teaching management system manages students, it is unable to adopt different monitoring methods according to the teaching methods of teachers to monitor the concentration of students, resulting in the lack of accuracy and pertinence of the concentration monitoring results.
[0005] Therefore, we propose an intelligent teaching management system based on an Internet classroom. Summary of the Invention
[0006] Aiming at the deficiencies of the existing technology, the purpose of the present invention is to provide an intelligent teaching management system based on an Internet classroom, and the present invention aims to improve the accuracy and pertinence of the teaching management system.
[0007] To achieve the above purpose, the present invention adopts the following technical solutions: An intelligent teaching management system based on an Internet classroom, including:
[0008] Data acquisition module: Perform spatial modeling on the location of the target student to obtain a student location space model, and perform real-time video stream acquisition on the student location space model to obtain target student acquisition data;
[0009] Model creation module: Analyze the video stream of the target student acquisition data, and map the student location space model according to the analysis result to obtain a location dynamic space model;
[0010] Action analysis module: Perform periodic action analysis on the target student in the location dynamic space model, and obtain the periodic blackboard writing teaching duration, monitoring period cumulative duration, first concentration analysis coefficient, and second concentration analysis coefficient according to the analysis result to obtain classroom action monitoring data;
[0011] Teaching management module: Perform classroom teaching management on the target student according to the classroom action monitoring data.
[0012] Furthermore, the acquisition of the target student acquisition data is specifically as follows:
[0013] Obtain the students in the Internet classroom and randomly select one student from the selected students as the target student;
[0014] Perform spatial marking on the spatial area where the target student is located, create a student position space model according to the marking result, and collect real-time video streams of the student position space model through monitoring devices to obtain spatial video stream data and obtain target student collection data;
[0015] Create a student position space model as follows:
[0016] Within the spatial area where the target student is located, mark the straight line at the front edge of the surface of the target student's desk as the first position feature straight line, mark the straight line at the left edge of the surface of the target student's desk as the second position feature straight line, and mark the straight line at the right edge of the surface of the target student's desk as the third position feature straight line;
[0017] Mark the ground in the spatial area where the target student is located as the first position feature plane, obtain the intersection point of the rearmost end of the target student's seat and the first position feature plane to get the first position feature point, and make a plane parallel to the first position feature straight line through the first position feature point to get the second position feature plane, and make a plane perpendicular to the second position feature plane through the first position feature straight line to get the third position feature plane;
[0018] Make a plane perpendicular to the third position feature plane through the second position feature straight line to get the fourth position feature plane, and make a plane parallel to the fourth position feature plane through the third position feature straight line to get the fifth position feature plane;
[0019] Obtain the height value of the target student to get the characteristic space height, and mark the plane at a perpendicular distance of the characteristic space height from the first position feature plane as the sixth position feature plane;
[0020] Mark the closed area enclosed by the first position feature plane, the second position feature plane, the third position feature plane, the fourth position feature plane, the fifth position feature plane and the sixth position feature plane as the student position space model.
[0021] Furthermore, create a position dynamic space model as follows:
[0022] Obtain the target student collection data, and obtain the student position space model and the spatial video stream data according to the target student collection data;
[0023] In the third position feature plane of the student position space model, arbitrarily select a model vertex as the coordinate origin, draw a line perpendicular to the first position feature plane through the coordinate origin to obtain the coordinate z-axis, mark the common edge of the first position feature plane and the third position feature plane as the coordinate x-axis, and in the first position feature plane, draw a line perpendicular to the coordinate x-axis through the coordinate origin to obtain the coordinate y-axis. Name the space coordinate system composed of the coordinate origin, the coordinate x-axis, the coordinate y-axis, and the coordinate z-axis as the three-dimensional coordinate system of the space model;
[0024] Obtain the cameras for shooting the student position space model images according to the spatial video stream data to get multiple position space cameras, and use the spatial registration algorithm to make the images collected by each position space camera align with the student position space model in the three-dimensional coordinate system of the space model;
[0025] Select the best-view camera for each surface pixel point in the student position space model according to the three-dimensional coordinate system of the space model, and use the best-view camera to perform pixel mapping on each pixel point to obtain the position dynamic space model.
[0026] Further, select the best-view camera as follows:
[0027] Obtain each surface pixel point in the student position space model, and arbitrarily select a sample surface pixel point from the obtained multiple surface pixel points, and select a sample space camera from the multiple position space cameras;
[0028] Obtain the coordinates of the sample surface pixel point P in the three-dimensional coordinate system of the space model as P = (x p , y p , z p ), obtain the straight line in the vertical direction of the sample surface pixel point P to get the normal vector n = (n x , n y , n z ), and obtain the coordinates of the sample space camera C i in the three-dimensional coordinate system of the space model as C i = (x ci , y ci , z ci );
[0029] Calculate the direction vector V i between the sample surface pixel point P and the sample space camera C i as follows:
[0030]
[0031] For the direction vector V iNormalize it, and the specific formula is as follows:
[0032]
[0033] Calculate the normal vector n and the direction vector V i The cosine value of the included angle between them, and the specific formula is as follows:
[0034]
[0035] For the sample surface pixel point P, obtain the cosine value of the included angle between each position space camera and the sample surface pixel point P respectively, and compare the obtained multiple cosine values of the included angles. If the cosine value of the included angle with the largest value is not blocked by an object, use the position space camera corresponding to the largest cosine value of the included angle as the best view camera for the sample surface pixel point P. If the cosine value of the included angle with the largest value is blocked by an object, use the position space camera corresponding to the sub-optimal cosine value of the included angle as the best view camera for the sample surface pixel point P.
[0036] Furthermore, obtain the classroom action monitoring data as follows:
[0037] Obtain the data collected by the target student, and based on the data collected by the target student, obtain the position dynamic space model;
[0038] In the position dynamic space model, mark the time point corresponding to the current moment as the end time point of the monitoring period. In the time period before the end time point of the monitoring period, mark a start time point of the monitoring period, and mark the time period between the start time point of the monitoring period and the end time point of the monitoring period as the student concentration monitoring period;
[0039] Set the head activity reference area of the target student in the position dynamic space model according to the classroom space layout to obtain the action reference monitoring area;
[0040] When the teacher gives a blackboard lecture, evaluate the concentration of the target student to obtain the first concentration analysis coefficient;
[0041] When the teacher does not give a blackboard lecture, evaluate the concentration of the target student to obtain the second concentration analysis coefficient;
[0042] Obtain the length of the time period corresponding to the teacher's blackboard lecture to get the cycle blackboard lecture duration, and obtain the duration of the student concentration monitoring period to get the cumulative monitoring period duration;
[0043] Define the cycle blackboard lecture duration, the cumulative monitoring period duration, the first concentration analysis coefficient, and the second concentration analysis coefficient as the classroom action monitoring data.
[0044] Further, obtain the action reference monitoring area as follows:
[0045] In the position dynamic space model, mark the position where the left eye of the target student is located as the first position reference point, mark the position where the right eye of the target student is located as the second position reference point, and obtain the midpoint of the line connecting the first position reference point and the second position reference point to get the third position reference point;
[0046] In the teaching space where the target student is located, mark the midpoint of the upper side edge of the classroom blackboard as the first purpose reference point, mark the midpoint of the left side edge of the classroom blackboard as the second purpose reference point, mark the midpoint of the right side edge of the classroom blackboard as the third purpose reference point, and mark the midpoint of the upper side edge of the classroom podium as the fourth purpose reference point;
[0047] Draw a line connecting the first position reference point and the second purpose reference point to obtain the first position reference line, mark the intersection point of the first position reference line and the third position feature plane as the first reference area feature point, draw a line connecting the second position reference point and the third purpose reference point to obtain the second position reference line, mark the intersection point of the second position reference line and the third position feature plane as the second reference area feature point, draw a line connecting the third position reference point and the first purpose reference point to obtain the third position reference line, mark the intersection point of the third position reference line and the third position feature plane as the third reference area feature point, draw a line connecting the third position reference point and the fourth purpose reference point to obtain the fourth position reference line, and mark the intersection point of the fourth position reference line and the third position feature plane as the fourth reference area feature point;
[0048] In the third position feature plane, draw a perpendicular line to the first position feature plane through the first reference area feature point to obtain the first area reference line, draw a perpendicular line to the first position feature plane through the second reference area feature point to obtain the second area reference line, draw a parallel line to the first position feature plane through the third reference area feature point to obtain the third area reference line, draw a parallel line to the first position feature plane through the fourth reference area feature point to obtain the fourth area reference line, and mark the closed area enclosed by the first area reference line, the second area reference line, the third area reference line, and the fourth area reference line as the action reference monitoring area.
[0049] Further, obtain the first concentration analysis coefficient as follows:
[0050] When the target student is in the student concentration monitoring period, obtain the intersection point of the first position reference line of the target student and the third position feature plane to get the first real-time position feature point, obtain the intersection point of the second position reference line of the target student and the third position feature plane to get the second real-time position feature point, obtain the intersection point of the third position reference line of the target student and the third position feature plane to get the third real-time position feature point, and obtain the intersection point of the fourth position reference line of the target student and the third position feature plane to get the fourth real-time position feature point;
[0051] During the student concentration monitoring period, obtain the time periods when the teacher teaches through blackboard writing, and sequentially mark the obtained multiple blackboard writing teaching time periods as the B1 blackboard writing teaching time period to the Ba blackboard writing teaching time period in chronological order;
[0052] During the B1 blackboard writing teaching time period, count the duration of the first real-time position feature point to the fourth real-time position feature point in the action reference monitoring area to obtain the first listening action duration to the fourth listening action duration, and calculate the average of the first listening action duration to the fourth listening action duration to obtain the listening action duration in the B1 time period;
[0053] Obtain the listening action durations in the B2 time period to the Ba time period corresponding to the B2 blackboard writing teaching time period to the Ba blackboard writing teaching time period respectively;
[0054] Obtain the numerical values of the time lengths of the B1 blackboard writing teaching time period to the Ba blackboard writing teaching time period to obtain the B1 blackboard writing teaching duration to the Ba blackboard writing teaching duration;
[0055] Calculate the first concentration analysis coefficient from the listening action durations in the B1 time period to the Ba time period and the B1 blackboard writing teaching duration to the Ba blackboard writing teaching duration;
[0056] Calculate the first concentration analysis coefficient, and the specific formula is as follows:
[0057]
[0058] Among them, Zzp1 is the first concentration analysis coefficient, Tkdi is the listening action duration in the Bi time period, Bssi is the Bi blackboard writing teaching duration, and a is the numerical value corresponding to the blackboard writing teaching time period.
[0059] Furthermore, obtain the second concentration analysis coefficient, specifically as follows:
[0060] Divide the student's concentration monitoring period into d concentration monitoring sub-periods with equal durations, select a sample concentration sub-period from the obtained several concentration monitoring sub-periods, and obtain the displacement speed of the head position corresponding to the sample concentration sub-period to get the sample displacement speed of the head position during the period;
[0061] Obtain the displacement speed of the head position corresponding to the interval period between every two consecutive monitoring time points respectively, get multiple displacement speeds of the head during the period, and calculate the average of the obtained multiple displacement speeds of the head during the period to get the displacement speed of the head position corresponding to the sample concentration sub-period;
[0062] Obtain the displacement speed of the head position corresponding to each concentration monitoring sub-period respectively, obtain the normal displacement speed interval. If the displacement speed of the head position is within the normal displacement speed interval, determine that the corresponding concentration monitoring sub-period is a normal head movement period; if the displacement speed of the head position is not within the normal displacement speed interval, determine that the corresponding concentration monitoring sub-period is an abnormal head movement period;
[0063] Count the number of normal head movement periods to get the number of target movement periods, and calculate the ratio of the number of target movement periods to d to get the second concentration analysis coefficient.
[0064] Furthermore, obtain the displacement speed of the head position during the sample period as follows:
[0065] Divide the student's concentration monitoring period into d concentration monitoring sub-periods with equal durations, and select a sample concentration sub-period from the obtained several concentration monitoring sub-periods;
[0066] Within the sample concentration sub-period, select several monitoring time points with equal time intervals, randomly select two consecutive monitoring time points from the obtained monitoring time points, name them the first sample monitoring time point and the second sample monitoring time point in chronological order, and obtain the numerical value of the interval duration between the first sample monitoring time point and the second sample monitoring time point to get the sample time point interval duration;
[0067] In the position dynamic space model, obtain the three-dimensional coordinates of the third position reference point at the first sample monitoring time point and the second sample monitoring time point through the three-dimensional coordinate system of the space model to get the first three-dimensional position coordinates (x w1 , y w1 , z w1 ) and the second three-dimensional position coordinates (x w2 , y w2 , z w2 );
[0068] Calculate the head displacement speed of the sample period by using the three-dimensional coordinates of the first position, the three-dimensional coordinates of the second position, and the time interval duration of the sample time points.
[0069] Calculate the head displacement speed of the sample period. The specific formula is as follows:
[0070]
[0071] Where, Vyb is the head displacement speed of the sample period, Tjg is the time interval duration of the sample time points, (x w1 , y w1 , z w1 ) are the three-dimensional coordinates of the first position, and (x w2 , y w2 , z w2 ) are the three-dimensional coordinates of the second position.
[0072] Furthermore, conduct classroom teaching management for the target student as follows:
[0073] Obtain classroom action monitoring data, and respectively obtain the cycle blackboard writing teaching duration, the cumulative monitoring cycle duration, the first concentration analysis coefficient, and the second concentration analysis coefficient according to the classroom action monitoring data;
[0074] Calculate the student concentration evaluation coefficient by using the cycle blackboard writing teaching duration, the cumulative monitoring cycle duration, the first concentration analysis coefficient, and the second concentration analysis coefficient;
[0075] Calculate the student concentration evaluation coefficient. The specific formula is as follows:
[0076]
[0077] Where, Zpg is the student concentration evaluation coefficient, Bks is the cycle blackboard writing teaching duration, Ljs is the cumulative monitoring cycle duration, Zzp1 is the first concentration analysis coefficient, and Zzp2 is the second concentration analysis coefficient;
[0078] Obtain the concentration evaluation benchmark interval. If the student concentration evaluation coefficient is within the concentration evaluation benchmark interval, it is determined that the target student's listening state is good. If the student concentration evaluation coefficient is not within the concentration evaluation benchmark interval, it is determined that the target student's listening state is poor, and a reminder for focused listening is given to the target.
[0079] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:
[0080] 1. The present invention analyzes the video stream of the collected data of the target students, maps the student position space model according to the analysis results to obtain a position dynamic space model, and analyzes the classroom actions of the students based on the position dynamic space model, which can effectively ensure the accuracy of the student action analysis, and further improve the accuracy of the classroom management level;
[0081] 2. The present invention monitors the concentration of students by adopting different action monitoring methods according to different teaching forms of teachers in the classroom, which can effectively ensure the pertinence of the concentration monitoring method, and further ensure the accuracy and pertinence of the concentration monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS
[0082] For the convenience of those skilled in the art to understand, the present invention will be further described below in conjunction with the drawings.
[0083] Figure 1 is the overall system block diagram of the present invention;
[0084] Figure 2 is the schematic diagram of the student position space model of the present invention;
[0085] Figure 3 is the schematic diagram of the three-dimensional coordinate system of the space model of the present invention. DETAILED DESCRIPTION OF THE INVENTION
[0086] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.
[0087] Embodiment 1
[0088] Please refer to Figure 1 , the present invention provides a technical solution: an intelligent teaching management system based on an Internet classroom, including a data acquisition module, a model creation module, an action analysis module, a teaching management module and a server. The data acquisition module, the model creation module, the action analysis module and the teaching management module are respectively connected to the server, and the server controls the data acquisition module, the model creation module, the action analysis module and the teaching management module respectively;
[0089] The data acquisition module performs spatial modeling on the position of the target students to obtain a student position space model, and performs real-time video stream acquisition on the student position space model to obtain the collected data of the target students;
[0090] Specifically as follows:
[0091] Obtain the students in the Internet classroom and randomly select one student from the selected students as the target student;
[0092] It should be noted here that:
[0093] In this application, the target student involved here is a sample of the objects served by the intelligent teaching management system.
[0094] Please refer to Figure 2 , perform spatial marking on the spatial area where the target student is located, and create a student position space model according to the marking results;
[0095] Specifically as follows:
[0096] Within the spatial area where the target student is located, mark the straight line at the front edge of the surface of the target student's desk as the first position feature straight line, mark the straight line at the left edge of the surface of the target student's desk as the second position feature straight line, and mark the straight line at the right edge of the surface of the target student's desk as the third position feature straight line;
[0097] Mark the ground in the spatial area where the target student is located as the first position feature plane, obtain the intersection point of the rearmost end of the target student's seat and the first position feature plane to get the first position feature point, make a plane parallel to the first position feature straight line through the first position feature point to get the second position feature plane, and make a plane perpendicular to the second position feature plane through the first position feature straight line to get the third position feature plane;
[0098] Make a plane perpendicular to the third position feature plane through the second position feature straight line to get the fourth position feature plane, and make a plane parallel to the fourth position feature plane through the third position feature straight line to get the fifth position feature plane;
[0099] Obtain the height value of the target student to get the characteristic space height, and mark the plane at a perpendicular distance of the characteristic space height from the first position feature plane as the sixth position feature plane;
[0100] Mark the enclosed area surrounded by the first position feature plane, the second position feature plane, the third position feature plane, the fourth position feature plane, the fifth position feature plane, and the sixth position feature plane as the student position space model; (bottom, top, front, left, right, back)
[0101] Collect real-time video stream data of the student position space model through a monitoring device to obtain spatial video stream data;
[0102] It should be noted here that:
[0103] In this application, the spatial video stream data involved herein includes video stream images collected by cameras at multiple different shooting angles, and the shooting images obtained by the cameras at multiple different shooting angles can achieve image coverage of the student position space model.
[0104] Define the student position space model and the spatial video stream data as the target student acquisition data.
[0105] The data acquisition module obtains the target student acquisition data and transports it to the model creation module, the action analysis module, and the limb movement module.
[0106] The model creation module performs video stream analysis on the target student acquisition data and textures the student position space model according to the analysis results to obtain the position dynamic space model.
[0107] Specifically as follows:
[0108] Obtain the target student acquisition data, and obtain the student position space model and the spatial video stream data according to the target student acquisition data.
[0109] Please refer to Figure 3 , in the third position feature plane of the student position space model, arbitrarily select a model vertex as the coordinate origin, draw a line perpendicular to the first position feature plane through the coordinate origin to obtain the coordinate z-axis, mark the common side of the first position feature plane and the third position feature plane as the coordinate x-axis, in the first position feature plane, draw a line perpendicular to the coordinate x-axis through the coordinate origin to obtain the coordinate y-axis, and name the space coordinate system composed of the coordinate origin, the coordinate x-axis, the coordinate y-axis, and the coordinate z-axis as the spatial model three-dimensional coordinate system.
[0110] It should be noted here that:
[0111] In this application, the spatial model three-dimensional coordinate system involved herein can extend to the space area outside the student position space model, that is, the three-dimensional coordinates of objects outside the student position space model in the spatial model three-dimensional coordinate system can be obtained.
[0112] Obtain the cameras that shoot the student position space model images according to the spatial video stream data to obtain multiple position space cameras, and use the spatial registration algorithm to align the images collected by each position space camera with the student position space model in the spatial model three-dimensional coordinate system.
[0113] It should be noted here that:
[0114] In this application, the spatial configuration algorithm involved herein is specifically Graphical Models.
[0115] Select the best - view camera for each surface pixel point in the student position space model according to the three - dimensional coordinate system of the space model, and use the best - view camera to perform pixel mapping on each pixel point to obtain the position dynamic space model;
[0116] Specifically as follows:
[0117] Obtain each surface pixel point in the student position space model, and arbitrarily select a sample surface pixel point from the obtained multiple surface pixel points, and select a sample space camera from multiple position - space cameras;
[0118] Obtain the coordinates of the sample surface pixel point P in the three - dimensional coordinate system of the space model as P=(x p ,y p ,z p ), obtain the straight line in the vertical direction of the sample surface pixel point P to get the normal vector n=(n x ,n y ,n z ), and obtain the coordinates of the sample space camera C i in the three - dimensional coordinate system of the space model as C i =(x ci ,y ci ,z ci );
[0119] Calculate the direction vector V i between the sample surface pixel point P and the sample space camera C i , specifically as follows:
[0120]
[0121] Normalize the direction vector V i , and the specific formula is as follows:
[0122]
[0123] Calculate the cosine value of the direction - angle between the normal vector n and the direction vector V i , and the specific formula is as follows:
[0124]
[0125] For the pixel point P on the sample surface, obtain the cosine value of the direction angle between each position-space camera and the pixel point P on the sample surface respectively, and compare the obtained multiple cosine values of the direction angles numerically. If the cosine value of the direction angle with the largest value is not blocked by an object, use the position-space camera corresponding to the largest cosine value of the direction angle as the best-view camera for the pixel point P on the sample surface. If the cosine value of the direction angle with the largest value is blocked by an object, use the position-space camera corresponding to the sub-optimal cosine value of the direction angle as the best-view camera for the pixel point P on the sample surface;
[0126] It should be noted here that:
[0127] In this application, the larger the cosine value of the direction angle, the better the camera view corresponding to the cosine value of the direction angle. The specific selection of the position-space camera corresponding to the sub-optimal cosine value of the direction angle involved here is that if the cosine value of the direction angle with the largest value is blocked by an object, then select the position-space camera corresponding to the second largest cosine value of the direction angle. If the cosine value of the second largest direction angle is blocked by an object, then select the position-space camera corresponding to the third largest cosine value of the direction angle, and so on;
[0128] In this application, the specific code implementation method for selecting the best-view camera here is as follows:
[0129]
[0130]
[0131] Repeat the process of selecting the best-view camera for the pixel point P on the sample surface, select the best-view camera for each surface pixel point in the student position-space model respectively, and use the video stream image obtained by the best-view camera to perform pixel mapping on each pixel point to obtain the position dynamic space model;
[0132] The model creation module obtains the position dynamic space model and transports it to the action analysis module and the limb movement module;
[0133] The action analysis module performs periodic action analysis on the target student in the position dynamic space model according to the target student's acquisition data, and obtains the periodic blackboard writing teaching duration, the cumulative monitoring period duration, the first concentration analysis coefficient, and the second concentration analysis coefficient according to the analysis results to obtain the classroom action monitoring data;
[0134] Specifically as follows:
[0135] Obtain the target student's acquisition data, and obtain the position dynamic space model according to the target student's acquisition data;
[0136] In the position dynamic space model, the time point corresponding to the current moment is marked as the end time point of the monitoring period. In the time period before the end time point of the monitoring period, a start time point of the monitoring period is marked, and the time period between the start time point and the end time point of the monitoring period is marked as the student concentration monitoring period;
[0137] It should be noted here that:
[0138] In this application, as the time value corresponding to the current moment changes, the end time point and the start time point of the monitoring period also change accordingly, thus realizing the dynamic update of the student concentration monitoring period;
[0139] In this application, the time value corresponding to the student concentration monitoring period involved here is specifically limited to 4 minutes. If the duration of a class is 40 minutes, then a class can be divided into 10 student concentration monitoring periods with equal durations;
[0140] According to the classroom space layout, the head activity reference area of the target student is set in the position dynamic space model to obtain the action reference monitoring area;
[0141] Specifically as follows:
[0142] In the position dynamic space model, the position where the left eye of the target student is located is marked as the first position reference point, the position where the right eye of the target student is located is marked as the second position reference point, and the midpoint of the line connecting the first position reference point and the second position reference point is obtained to get the third position reference point;
[0143] It should be noted here that:
[0144] In this application, the first position reference point, the second position reference point, and the third position reference point involved here are measured when the target student is in the standard sitting posture;
[0145] The standard sitting posture involved here is the student sitting posture standard set in the school teaching management regulations.
[0146] In the teaching space where the target student is located, the midpoint of the upper side edge of the classroom blackboard is marked as the first purpose reference point, the midpoint of the left side edge of the classroom blackboard is marked as the second purpose reference point, the midpoint of the right side edge of the classroom blackboard is marked as the third purpose reference point, and the midpoint of the upper side edge of the classroom podium is marked as the fourth purpose reference point;
[0147] Connect the first position reference point and the second target reference point to obtain the first position reference connection line. Mark the intersection point of the first position reference connection line and the third position feature plane as the first reference area feature point. Connect the second position reference point and the third target reference point to obtain the second position reference connection line. Mark the intersection point of the second position reference connection line and the third position feature plane as the second reference area feature point. Connect the third position reference point and the first target reference point to obtain the third position reference connection line. Mark the intersection point of the third position reference connection line and the third position feature plane as the third reference area feature point. Connect the third position reference point and the fourth target reference point to obtain the fourth position reference connection line. Mark the intersection point of the fourth position reference connection line and the third position feature plane as the fourth reference area feature point;
[0148] In the third position feature plane, draw a perpendicular line to the first position feature plane through the first reference area feature point to obtain the first area reference line. Draw a perpendicular line to the first position feature plane through the second reference area feature point to obtain the second area reference line. Draw a parallel line to the first position feature plane through the third reference area feature point to obtain the third area reference line. Draw a parallel line to the first position feature plane through the fourth reference area feature point to obtain the fourth area reference line. Mark the closed area enclosed by the first area reference line, the second area reference line, the third area reference line, and the fourth area reference line as the action reference monitoring area;
[0149] When the target student is in the student concentration monitoring period, obtain the intersection point of the first position reference connection line of the target student and the third position feature plane to obtain the first real-time position feature point. Obtain the intersection point of the second position reference connection line of the target student and the third position feature plane to obtain the second real-time position feature point. Obtain the intersection point of the third position reference connection line of the target student and the third position feature plane to obtain the third real-time position feature point. Obtain the intersection point of the fourth position reference connection line of the target student and the third position feature plane to obtain the fourth real-time position feature point;
[0150] During the student concentration monitoring period, when the teacher is giving a blackboard lecture, conduct a concentration assessment on the target student to obtain the first concentration analysis coefficient;
[0151] Specifically as follows:
[0152] During the student concentration monitoring period, obtain the periods when the teacher gives a blackboard lecture, and sequentially mark the multiple obtained blackboard lecture periods as the B1 blackboard lecture period to the Ba blackboard lecture period in chronological order;
[0153] It should be noted here that:
[0154] In this application, B involved here is the identifier corresponding to the blackboard teaching period, a is the numerical value corresponding to the blackboard teaching period, and a is an integer greater than 0.
[0155] During the B1 blackboard teaching period, the durations of the first to fourth real-time position feature points in the action benchmark monitoring area are counted to obtain the first to fourth listening action durations, and the average of the first to fourth listening action durations is calculated to obtain the listening action duration in the B1 period.
[0156] The listening action durations corresponding to the B2 to Ba blackboard teaching periods are obtained respectively to obtain the listening action duration in the B2 period to the listening action duration in the Ba period.
[0157] The numerical values of the lengths of the B1 to Ba blackboard teaching periods are obtained to obtain the blackboard teaching duration of B1 to the blackboard teaching duration of Ba.
[0158] The listening action durations from the B1 period to the Ba period and the blackboard teaching durations from the B1 blackboard teaching duration to the Ba blackboard teaching duration are calculated to obtain the first concentration analysis coefficient.
[0159] The first concentration analysis coefficient is calculated, and the specific formula is as follows:
[0160]
[0161] Among them, Zzp1 is the first concentration analysis coefficient, Tkdi is the listening action duration in the Bi period, Bssi is the blackboard teaching duration in the Bi period, and a is the numerical value corresponding to the blackboard teaching period.
[0162] It should be noted here that:
[0163] In this application, the listening action duration in the Bi period involved here can be any one of the listening action durations from the B1 period to the Ba period, and the blackboard teaching duration in the Bi period involved here can be any one of the blackboard teaching durations from the B1 blackboard teaching duration to the Ba blackboard teaching duration.
[0164] In a specific implementation, the following experimental data are measured:
[0165] Bss1 is 80s, Bss2 is 90s, Bss3 is 75s, Zzp1 is 71s, Zzp2 is 82s, Zzp3 is 65s, then the calculated first concentration analysis coefficient is 0.89.
[0166] During the student concentration monitoring period, when the teacher is not giving a blackboard lecture, the concentration of the target student is evaluated to obtain the second concentration analysis coefficient;
[0167] Specifically as follows:
[0168] The student concentration monitoring period is divided into d concentration monitoring sub-periods of equal duration, and a sample concentration sub-period is selected from the obtained several concentration monitoring sub-periods;
[0169] It should be noted here that:
[0170] In this application, the time length corresponding to the concentration monitoring sub-period involved here is specifically limited to four seconds;
[0171] In this application, d involved here is the numerical value corresponding to the number of concentration monitoring sub-periods, and d is an integer greater than 0;
[0172] Within the sample concentration sub-period, several monitoring time points with equal time intervals are selected, and any two consecutive monitoring time points are randomly selected from the obtained monitoring time points, and they are named the first sample monitoring time point and the second sample monitoring time point in chronological order, and the interval duration between the first sample monitoring time point and the second sample monitoring time point is numerically obtained to obtain the sample time point interval duration;
[0173] It should be noted here that:
[0174] In this application, the first sample monitoring time point involved here is before the second sample monitoring time point.
[0175] In the position dynamic space model, the three-dimensional coordinates of the third position reference point at the first sample monitoring time point and the second sample monitoring time point are obtained through the three-dimensional coordinate system of the space model, and the first position three-dimensional coordinates (x w1 , y w1 , z w1 ) and the second position three-dimensional coordinates (x w2 , y w2 , z w2 ) are obtained;
[0176] The first position three-dimensional coordinates, the second position three-dimensional coordinates, and the sample time point interval duration are calculated to obtain the displacement speed of the head part in the sample period;
[0177] The displacement speed of the head part in the sample period is calculated, and the specific formula is as follows:
[0178]
[0179] Among them, Vyb is the displacement speed of the head part in the sample period, Tjg is the sample time point interval duration, (xw1 , y w1 , z w1 ) is the three-dimensional coordinate of the first position, (x w2 , y w2 , z w2 ) is the three-dimensional coordinate of the second position;
[0180] In a specific implementation, there are the following experimental data:
[0181] The three-dimensional coordinate of the first position is (2.8, 3.0, 4.2), the three-dimensional coordinate of the first position is (5.7, 6.7, 9.9), and the time interval between sample time points is 1.2 s. Then, the displacement speed of the head in the time period can be calculated to be 6.15 cm / s.
[0182] Repeat for the displacement speed of the head in the sample time period, and obtain the displacement speed of the head in the time period corresponding to the interval between every two consecutive monitoring time points respectively, to obtain multiple displacement speeds of the head in the time period, and calculate the average of the obtained multiple displacement speeds of the head in the time period to obtain the displacement speed of the head in the period corresponding to the sample concentration sub-period;
[0183] Obtain the displacement speed of the head in the period corresponding to each concentration monitoring sub-period respectively, obtain the normal displacement speed interval. If the displacement speed of the head in the period is within the normal displacement speed interval, it is determined that the corresponding concentration monitoring sub-period is a normal head movement period. If the displacement speed of the head in the period is not within the normal displacement speed interval, it is determined that the corresponding concentration monitoring sub-period is an abnormal head movement period;
[0184] It should be noted here that:
[0185] In this application, the normal head movement period involved here includes the interval boundaries corresponding to the normal displacement speed interval.
[0186] Count the number of normal head movement periods to obtain the number of target movement periods, and calculate the ratio of the number of target movement periods to d to obtain the second concentration analysis coefficient;
[0187] It should be noted here that:
[0188] It is measured through experiments that the normal displacement speed interval involved here is set to [3, 7];
[0189] The specific experimental process is as follows:
[0190] Number of students: 300
[0191] Each student conducts 3 experiments, and the duration of the experimental period is equal to the duration corresponding to the concentration monitoring sub-period.
[0192] Average head displacement speed (cm / s): The average speed of all students in all sub - cycles is 5.0 cm / s
[0193] Standard deviation: The standard deviation of the speed of all students in all sub - cycles is 2.0 cm / s
[0194] Based on these data, we can set the normal displacement speed range as:
[0195] Lower limit: Average speed - standard deviation = 5.0 cm / s - 2.0 cm / s = 3.0 cm / s
[0196] Upper limit: Average speed + standard deviation = 5.0 cm / s + 2.0 cm / s = 7.0 cm / s.
[0197] During the student concentration monitoring period, obtain the length of the time period corresponding to the teacher's blackboard writing teaching, get the blackboard writing teaching duration of the period, obtain the cumulative duration of the monitoring period for student concentration;
[0198] Define the blackboard writing teaching duration of the period, the cumulative duration of the monitoring period, the first concentration analysis coefficient, and the second concentration analysis coefficient as classroom action monitoring data;
[0199] The teaching management module conducts classroom teaching management on the target students according to the classroom action monitoring data;
[0200] Specifically as follows:
[0201] Obtain the classroom action monitoring data, and respectively obtain the blackboard writing teaching duration of the period, the cumulative duration of the monitoring period, the first concentration analysis coefficient, and the second concentration analysis coefficient according to the classroom action monitoring data;
[0202] Obtain the student concentration evaluation coefficient by calculating the blackboard writing teaching duration of the period, the cumulative duration of the monitoring period, the first concentration analysis coefficient, and the second concentration analysis coefficient;
[0203] Calculate the student concentration evaluation coefficient, and the specific formula is as follows:
[0204]
[0205] Among them, Zpg is the student concentration evaluation coefficient, Bks is the blackboard writing teaching duration of the period, Ljs is the cumulative duration of the monitoring period, Zzp1 is the first concentration analysis coefficient, and Zzp2 is the second concentration analysis coefficient;
[0206] It should be noted here that:
[0207] Here, the ratio of the blackboard writing teaching duration and the non-blackboard writing duration within the student concentration monitoring period to the period duration is used as a proportionality coefficient, and in combination with the first concentration analysis coefficient and the second concentration analysis coefficient, the concentration of the target student in the student concentration monitoring period is jointly monitored. It can adopt different action monitoring methods for students according to different teaching forms of teachers in the classroom, which can effectively ensure the pertinence of the concentration monitoring method, and then ensure the accuracy and pertinence of the concentration monitoring results;
[0208] On the other hand, blackboard writing teaching usually involves more knowledge explanation and thinking guidance, while non-blackboard writing moments may include discussions, experiments, or self-study, etc. This distinction enables the monitoring to more accurately reflect the concentration of students in different teaching links.
[0209] It should be noted here that:
[0210] In specific implementation, the following experimental data are measured. The blackboard writing teaching duration of the period is 23 minutes, the cumulative duration of the monitoring period is 40 minutes, the first concentration analysis coefficient is 0.71, and the second concentration analysis coefficient is 0.61. The student concentration evaluation coefficient can be calculated to be 0.66725;
[0211] Obtain the concentration evaluation benchmark interval. If the student concentration evaluation coefficient is within the concentration evaluation benchmark interval, it is determined that the target student's listening state is good. If the student concentration evaluation coefficient is outside the concentration evaluation benchmark interval, it is determined that the target student's listening state is poor, and a reminder for focused listening is given to the target.
[0212] It should be noted here that:
[0213] The good listening state involved here includes the interval boundaries corresponding to the concentration evaluation benchmark interval.
[0214] Through experiments, the concentration evaluation benchmark interval is [0.6408, 1.0];
[0215] The specific experimental process is as follows: Given that the maximum value of the calculation result of the student concentration evaluation coefficient is 1.0, so the right endpoint of the interval of the concentration evaluation benchmark interval is 1.0;
[0216] In specific implementation, 300 experimental students known to be in a focused listening state are selected from historical detection data. Through experiments, the average value Pjz of the student concentration evaluation coefficients corresponding to the 300 selected experimental students is 0.72, and the standard deviation Bzc of the student concentration evaluation coefficients corresponding to the 300 selected experimental students is 0.11. Through the formula Xx = Pjz×(1 - Bzc), the left endpoint Xx of the interval of the concentration evaluation benchmark interval is calculated to be 0.6408;
[0217] In this application, if there are corresponding calculation formulas, the above calculation formulas are all calculated by taking the numerical values without dimensions. For the coefficients such as the weight coefficient and the proportionality coefficient existing in the formula, the magnitudes set are for obtaining a result value by quantifying each parameter. Regarding the magnitudes of the weight coefficient and the proportionality coefficient, as long as the proportional relationship between the parameters and the result value is not affected.
[0218] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to only the specific implementation manners. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification to better explain the principle and practical application of the present invention, so that those skilled in the art can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.
Claims
1. An intelligent teaching management system based on Internet classrooms, characterized in that Including: Data acquisition module: Perform spatial modeling on the location of the target student to obtain a student location space model, and collect real-time video streams for the student location space model to obtain target student acquisition data; Model creation module: Analyze the video stream of the target student acquisition data, and texture the student location space model according to the analysis results to obtain a location dynamic space model; Action analysis module: Perform periodic action analysis on the target student in the location dynamic space model, and obtain the periodic blackboard writing teaching duration, monitoring cycle cumulative duration, first concentration analysis coefficient, and second concentration analysis coefficient according to the analysis results to obtain classroom action monitoring data; Teaching management module: Conduct classroom teaching management on the target student according to the classroom action monitoring data.
2. The intelligent teaching management system based on Internet classroom according to claim 1, characterized in that, Obtain the target student acquisition data as follows: Obtain the students in the Internet classroom, and arbitrarily select one student as the target student according to the acquisition results; Spatially mark the spatial area where the target student is located, create a student location space model according to the marking results, and collect real-time video streams for the student location space model through monitoring devices to obtain spatial video stream data, and obtain target student acquisition data; Specifically as follows: Within the spatial area where the target student is located, mark the first position feature plane to the fifth position feature plane respectively; Obtain the height value of the target student to obtain the feature space height, and mark the plane with a perpendicular distance of the feature space height from the first position feature plane as the sixth position feature plane; Mark the enclosed area surrounded by the first position feature plane to the sixth position feature plane as the student location space model.
3. The intelligent teaching management system based on the Internet classroom according to claim 1 is characterized in that, Create the location dynamic space model as follows: Obtain the target student acquisition data, and obtain the student location space model and spatial video stream data according to the target student acquisition data; Create a spatial coordinate system in the student location space model to obtain a spatial model three-dimensional coordinate system; Obtain the cameras that capture the images of the student location space model according to the spatial video stream data to obtain multiple location space cameras, and use the spatial registration algorithm to align the images collected by each location space camera with the student location space model in the spatial model three-dimensional coordinate system; Select the best-view camera for each surface pixel point in the student location space model according to the spatial model three-dimensional coordinate system, and use the best-view camera to perform pixel mapping on each pixel point to obtain a location dynamic space model.
4. The intelligent teaching management system based on Internet classroom according to claim 3, wherein, Select the best-view camera as follows: Select a sample surface pixel point in the student location space model, and select a sample spatial camera from multiple location space cameras; The coordinates of the pixel point P on the sample surface in the three-dimensional coordinate system of the spatial model are P = (x p , y p , z p ). The straight line in the vertical direction of the pixel point P on the sample surface is obtained, and the normal vector n = (n x , n y , n z ) is obtained. The coordinates of the sample space camera C i in the three-dimensional coordinate system of the spatial model are C i = (x ci , y ci , z ci ); Calculate the direction vector V between the pixel point P on the sample surface and the camera C in the sample space i as follows: i Specifically, For the direction vector V i perform unitization, and the specific formula is as follows: Calculate the cosine value of the directional angle between the normal vector n and the direction vector V i The specific formula is as follows: For the pixel point P on the sample surface, obtain the cosine values of the direction angles between each position-space camera and the pixel point P on the sample surface respectively, and compare the obtained multiple cosine values of the direction angles numerically. If the cosine value of the direction angle with the largest value is not blocked by an object, use the position-space camera corresponding to the largest cosine value of the direction angle as the best-view camera for the pixel point P on the sample surface. If the cosine value of the direction angle with the largest value is blocked by an object, use the position-space camera corresponding to the sub-optimal cosine value of the direction angle as the best-view camera for the pixel point P on the sample surface.
5. A smart teaching management system based on an Internet classroom according to claim 1, characterized in that, Obtain the classroom action monitoring data as follows: Obtain the data collected by the target student, and based on the data collected by the target student, obtain the position dynamic space model; During the class period of the target student, mark a monitoring period for the student's concentration; Set the head movement reference area of the target student in the position dynamic space model according to the classroom space layout to obtain the action reference monitoring area; When the teacher gives a lecture by writing on the blackboard, evaluate the concentration of the target student to obtain the first concentration analysis coefficient; When the teacher does not give a lecture by writing on the blackboard, evaluate the concentration of the target student to obtain the second concentration analysis coefficient; Obtain the length of the period corresponding to the teacher's lecture by writing on the blackboard to obtain the periodic lecture duration by writing on the blackboard, and obtain the cumulative duration of the monitoring period for the student's concentration to obtain the cumulative monitoring period duration; Define the periodic lecture duration by writing on the blackboard, the cumulative monitoring period duration, the first concentration analysis coefficient, and the second concentration analysis coefficient as the classroom action monitoring data.
6. The intelligent teaching management system based on the Internet classroom according to claim 5, characterized in that, Obtain the action reference monitoring area as follows: Mark the first position reference point to the third position reference point on the face of the target student in the position dynamic space model; Mark the first purpose reference point to the fourth purpose reference point in the teaching space where the target student is located; Obtain the intersection points of the interaction lines connecting the first position reference point to the third position reference point and the first purpose reference point to the fourth purpose reference point with the third position feature plane to obtain the first reference area feature point to the fourth reference area feature point; In the third position feature plane, draw the first area reference line to the fourth area reference line through the first reference area feature point to the fourth reference area feature point respectively, and mark the closed area enclosed by the first area reference line to the fourth area reference line as the action reference monitoring area.
7. The intelligent teaching management system based on Internet classroom according to claim 5, wherein The obtaining of the first concentration analysis coefficient is as follows: When the target student is in the monitoring period of the student's concentration, obtain the intersection point of the first position reference connection line and the third position feature plane to obtain the first real-time position feature point, obtain the intersection point of the second position reference connection line and the third position feature plane to obtain the second real-time position feature point, obtain the intersection point of the third position reference connection line and the third position feature plane to obtain the third real-time position feature point, and obtain the intersection point of the fourth position reference connection line and the third position feature plane to obtain the fourth real-time position feature point; During the monitoring period of the student's concentration, obtain the periods when the teacher gives a lecture by writing on the blackboard, and mark the obtained multiple lecture periods by writing on the blackboard as the B1 lecture period by writing on the blackboard to the Ba lecture period by writing on the blackboard in chronological order; During the B1 blackboard writing teaching period, count the duration of the first to fourth real-time position feature points in the action benchmark monitoring area to obtain the first to fourth listening action durations. Calculate the average of the first to fourth listening action durations to obtain the listening action duration in the B1 period. Obtain the listening action durations in the B2 to Ba blackboard writing teaching periods, namely, the listening action duration in the B2 period to the listening action duration in the Ba period. Obtain the numerical values of the lengths of the B1 to Ba blackboard writing teaching periods, namely, the B1 blackboard writing teaching duration to the Ba blackboard writing teaching duration. Calculate the first concentration analysis coefficient from the listening action durations in the B1 to Ba periods and the B1 to Ba blackboard writing teaching durations. Calculate the first concentration analysis coefficient, and the specific formula is as follows: Among them, Zzp1 is the first concentration analysis coefficient, Tkdi is the listening action duration in the Bi period, Bssi is the Bi blackboard writing teaching duration, and a is the numerical value corresponding to the number of blackboard writing teaching periods.
8. An intelligent teaching management system based on Internet classrooms according to claim 5, characterized in that, Obtain the second concentration analysis coefficient, specifically as follows: Divide the student concentration monitoring period into d concentration monitoring sub-periods of equal duration, and select a sample concentration sub-period from the obtained several concentration monitoring sub-periods. Obtain the head displacement speed corresponding to the sample concentration sub-period to get the sample time period head displacement speed. Obtain the head displacement speeds corresponding to the time intervals between every two consecutive monitoring time points respectively, to get multiple head time period displacement speeds, and calculate the average of the obtained multiple head time period displacement speeds to get the cycle head displacement speed corresponding to the sample concentration sub-period. Obtain the normal displacement speed range by obtaining the cycle head displacement speeds corresponding to each concentration monitoring sub-period. If the cycle head displacement speed is within the normal displacement speed range, determine that the corresponding concentration monitoring sub-period is a normal head movement cycle; if the cycle head displacement speed is not within the normal displacement speed range, determine that the corresponding concentration monitoring sub-period is an abnormal head movement cycle. Count the number of normal head movement cycles to obtain the target movement cycle number, and calculate the ratio of the target movement cycle number to d to obtain the second concentration analysis coefficient.
9. The intelligent teaching management system based on the Internet classroom according to claim 8, wherein, Obtain the sample time period head displacement speed, specifically as follows: In the sample concentration sub-period, select several monitoring time points with equal time intervals, and arbitrarily select two consecutive monitoring time points from the obtained monitoring time points, and name them the first sample monitoring time point and the second sample monitoring time point in chronological order. Obtain the numerical value of the time interval duration between the first sample monitoring time point and the second sample monitoring time point to get the sample time point interval duration. In the position dynamic space model, the three-dimensional coordinates of the third position reference point at the first sample monitoring time point and the second sample monitoring time point are obtained through the three-dimensional coordinate system of the space model, and the first position three-dimensional coordinates (x w1 , y w1 , z w1 ) and the second position three-dimensional coordinates (x w2 , y w2 , z w2 ) are obtained; The three-dimensional coordinates of the first position (x w1 , y w1 , z w1 ), the three-dimensional coordinates of the second position (x w2 , y w2 , z w2 ), and the time interval duration Tjg of the sample time points are used to calculate the head displacement velocity Vyb of the sample period:
10. A smart teaching management system based on Internet classrooms according to claim 1, characterized in that, Conduct classroom teaching management for the target students, specifically as follows: Obtain classroom action monitoring data, and respectively obtain the cycle blackboard writing teaching duration, the cumulative duration of the monitoring cycle, the first concentration analysis coefficient, and the second concentration analysis coefficient according to the classroom action monitoring data; Obtain the student concentration evaluation coefficient Zpg by calculating the cycle blackboard writing teaching duration Bks, the cumulative duration of the monitoring cycle Ljs, the first concentration analysis coefficient Zzp1, and the second concentration analysis coefficient Zzp2. The specific formula is as follows: Obtain the concentration evaluation benchmark interval. If the student concentration evaluation coefficient is within the concentration evaluation benchmark interval, it is determined that the target student's listening state is good; If the student concentration evaluation coefficient is within the concentration evaluation benchmark interval, it is determined that the target student's listening state is poor, and a reminder for focused listening is given to the target.