CN110443226A - A kind of student's method for evaluating state and system based on gesture recognition - Google Patents
A kind of student's method for evaluating state and system based on gesture recognition Download PDFInfo
- Publication number
- CN110443226A CN110443226A CN201910758393.5A CN201910758393A CN110443226A CN 110443226 A CN110443226 A CN 110443226A CN 201910758393 A CN201910758393 A CN 201910758393A CN 110443226 A CN110443226 A CN 110443226A
- Authority
- CN
- China
- Prior art keywords
- student
- video clip
- posture
- classroom
- indicate
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000011156 evaluation Methods 0.000 claims abstract description 20
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 6
- 230000036544 posture Effects 0.000 claims description 102
- 238000012549 training Methods 0.000 claims description 16
- 239000011159 matrix material Substances 0.000 claims description 15
- 230000003993 interaction Effects 0.000 claims description 10
- 210000003127 knee Anatomy 0.000 claims description 10
- 230000033001 locomotion Effects 0.000 claims description 8
- 230000001133 acceleration Effects 0.000 claims description 7
- 239000012141 concentrate Substances 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 210000001699 lower leg Anatomy 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- 238000007637 random forest analysis Methods 0.000 claims description 4
- 230000015572 biosynthetic process Effects 0.000 claims description 2
- 230000007787 long-term memory Effects 0.000 claims description 2
- 229910017435 S2 In Inorganic materials 0.000 claims 2
- 210000001217 buttock Anatomy 0.000 claims 2
- 238000012552 review Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 6
- SBNFWQZLDJGRLK-UHFFFAOYSA-N phenothrin Chemical compound CC1(C)C(C=C(C)C)C1C(=O)OCC1=CC=CC(OC=2C=CC=CC=2)=C1 SBNFWQZLDJGRLK-UHFFFAOYSA-N 0.000 description 4
- 238000009795 derivation Methods 0.000 description 3
- 206010057315 Daydreaming Diseases 0.000 description 2
- 206010041349 Somnolence Diseases 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000006073 displacement reaction Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000010754 BS 2869 Class F Substances 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000035800 maturation Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- 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)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Multimedia (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Technology (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of student's method for evaluating state and system based on gesture recognition.This method comprises: obtaining the classroom video and teachers' instruction video of all students on classroom in real time;Compartment of terrain intercepts multiple video clips from the video of classroom;Each video clip is associated with a time tag;It is partitioned into the single video clip of all students in video clip;For any single video clip, the personal information of single video clip middle school student is identified using face recognition algorithms, identify that the posture of single video clip middle school student exports all corresponding time tags of posture for not meeting classroom requirement in the classroom performance evaluation and set of each student's posture time of each student based on gesture recognition model.The posture scientifically integrated in all video clips of student carries out classroom performance evaluation to student, and record student does not meet classroom and requires the corresponding time tag of posture, reviews convenient for student's specific aim after class and omit knowledge point, and teacher can also improve the mode of giving lessons according to this.
Description
Technical field
The present invention relates to intellectual education field, more particularly to a kind of student's method for evaluating state based on gesture recognition and
System.
Background technique
With the maturation of image technique, artificial intelligence technology is widely applied to classroom teaching.University curriculum it is heavy and
Class hour is limited, and teacher needs to be completed in a relatively short time teaching task.And university curriculum student play truant, attend class drowsiness, it is absent-minded
The phenomenon that happen occasionally.In order to guarantee that Classroom Teaching, College Teachers often interact, such as requires student to answer instantly and ask
Topic maintains classroom discipline, reminds drowsiness, absent-minded student, and record student classroom performance undoubtedly increases difficulty of giving lessons to teacher
Degree, also reduces the continuity that classroom is given lessons, influences teaching quality.Therefore, how automatically to the classroom of each student show into
The accurately evaluation of row science, has great importance.
Summary of the invention
The present invention is directed at least solve the technical problems existing in the prior art, especially innovatively propose a kind of based on appearance
The student's method for evaluating state and system of state identification.
In order to realize above-mentioned purpose of the invention, according to the first aspect of the invention, the present invention provides one kind to be based on
Student's method for evaluating state of gesture recognition, comprising:
Step S1 obtains the classroom video and teachers' instruction video of all students on classroom in real time, stores teachers' instruction
Video;
Compartment of terrain intercepts multiple video clips from the classroom video;
Each video clip is associated with the time tag for recording the video clip actual photographed time;
Each video clip is handled as follows in step S2:
It is partitioned into the single video clip of all students in video clip;
For any single video clip, of the single video clip middle school student is identified using face recognition algorithms
People's information identifies the posture of the single video clip middle school student based on gesture recognition model, by the personal information and appearance
State is associated;
Step S3 utilizes the personal information obtained from all video clips and appearance associated with the personal information
State constructs the posture time set of each student, if the posture time collection of k-th of student is combined into Uk={ [sk1,sk2,...,
ski...,skN], [T1,T2,...,Ti,...,TN]};
Wherein, 1≤k≤n, n indicate the total number of persons of student, are positive integer;N indicates the total quantity of video clip, is positive whole
Number;skiIndicate posture of k-th of student in i-th of video clip;TiIndicate the time tag of i-th of video clip, 1≤i
≤N;skiWith TiIt corresponds;
The classroom that posture time set based on all students obtains each student shows evaluation;
Step S4 is exported and all is not inconsistent in the classroom performance evaluation and set of each student's posture time of each student
Close the corresponding time tag of posture that classroom requires.
Above-mentioned technical proposal has the beneficial effect that this method acquires classroom video and teachers' instruction video in real time, from class
Each student is identified in hall video in the posture of each video clip, scientifically integrates the posture in all video clips of student to
It is raw to carry out classroom performance evaluation, while recording student and not meeting the corresponding time tag of classroom requirement posture, in this way convenient for after class
Student targetedly reviews the knowledge point of omission in conjunction with these time tags and teachers' instruction video, and teacher can also check
The corresponding video content of giving lessons of the time tag that more student does not pay attention to the class conscientiously, improves mode of giving lessons, preferably improves the quality of teaching.
In order to realize above-mentioned purpose of the invention, according to the second aspect of the invention, the present invention provides a kind of students
Classroom state evaluation system, the first camera including all students on shooting classroom, shoots teachers' instruction content on classroom
Second camera and server;
The server receives the student classroom video of the first camera output and the teachers' instruction of second camera output
Video, and commented according to the classroom performance that student's method for evaluating state of the present invention based on gesture recognition obtains each student
Valence, and the classroom that do not meet for saving student requires the corresponding time tag of posture and storage teachers' instruction video.
Above-mentioned technical proposal has the beneficial effect that the system acquires classroom video and teachers' instruction video in real time, saves
The classroom that do not meet of student requires the corresponding time tag of posture and teachers' instruction video, and each is identified from the video of classroom
The raw posture in each video clip, the posture scientifically integrated in all video clips of student comment student's progress classroom performance
Valence, while recording student and not meeting the corresponding time tag of classroom requirement posture, these times are combined in this way convenient for student after class
Label and teachers' instruction video targetedly review the knowledge point of omission, and teacher can also check that more student does not listen conscientiously
The corresponding video content of giving lessons of the time tag said, improves mode of giving lessons, preferably improves the quality of teaching.
Detailed description of the invention
Fig. 1 is the process signal of student's method for evaluating state in the embodiment of the invention based on gesture recognition
Figure;
Fig. 2 is three pose gesture schematic diagram in the embodiment of the invention;
Fig. 3 is the structural schematic diagram of gesture recognition model in the embodiment of the invention;
Fig. 4 is system layout schematic diagram in the embodiment of the invention;
Fig. 5 is system connection schematic diagram in the embodiment of the invention.
Appended drawing reference:
The head a;B neck;The left shoulder of c;The back d;The left elbow of e;F coccyx root;G left hand;The left stern of h;The left knee of u;V right crus of diaphragm;W is left
Foot;The right knee of m;The right stern of z;The o right hand;The right elbow of p;The right shoulder of q;1 first camera;2 second cameras;3 servers;4 student terminals are set
It is standby.
Specific embodiment
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached
The embodiment of figure description is exemplary, and for explaining only the invention, and is not considered as limiting the invention.
In the description of the present invention, it is to be understood that, term " longitudinal direction ", " transverse direction ", "upper", "lower", "front", "rear",
The orientation or positional relationship of the instructions such as "left", "right", "vertical", "horizontal", "top", "bottom" "inner", "outside" is based on attached drawing institute
The orientation or positional relationship shown, is merely for convenience of description of the present invention and simplification of the description, rather than the dress of indication or suggestion meaning
It sets or element must have a particular orientation, be constructed and operated in a specific orientation, therefore should not be understood as to limit of the invention
System.
In the description of the present invention, unless otherwise specified and limited, it should be noted that term " installation ", " connected ",
" connection " shall be understood in a broad sense, for example, it may be mechanical connection or electrical connection, the connection being also possible to inside two elements can
, can also indirectly connected through an intermediary, for the ordinary skill in the art to be to be connected directly, it can basis
Concrete condition understands the concrete meaning of above-mentioned term.
The invention discloses a kind of student's method for evaluating state based on gesture recognition, in a preferred embodiment,
The flow diagram of this method is as shown in Figure 1, specifically include:
Step S1 obtains the classroom video and teachers' instruction video of all students on classroom in real time, stores teachers' instruction
Video;
Compartment of terrain intercepts multiple video clips from the video of classroom;
Each video clip is associated with the time tag of a record video clip actual photographed time;
Each video clip is handled as follows in step S2:
It is partitioned into the single video clip of all students in video clip;
For any single video clip, identify that the personal of single video clip middle school student is believed using face recognition algorithms
Breath, the posture of single video clip middle school student is identified based on gesture recognition model, and personal information is associated with posture;
Step S3 utilizes the personal information obtained from all video clips and appearance associated with the personal information
State constructs the posture time set of each student, if the posture time collection of k-th of student is combined into Uk={ [sk1,sk2,...,
ski...,skN], [T1,T2,...,Ti,...,TN]};
Wherein, 1≤k≤n, n indicate the total number of persons of student, are positive integer;N indicates the total quantity of video clip, is positive whole
Number;skiIndicate posture of k-th of student in i-th of video clip;TiIndicate the time tag of i-th of video clip, 1≤i
≤N;skiWith TiIt corresponds;
The classroom that posture time set based on all students obtains each student shows evaluation;
Step S4 is exported and all is not inconsistent in the classroom performance evaluation and set of each student's posture time of each student
Close the corresponding time tag of posture that classroom requires.
In the present embodiment, multiple piece of video with equal or unequal time interval can be intercepted from the video of classroom
Section, the interval time of equal time distances can be 3-7 seconds, it is preferred that may be selected to be 5 seconds.
In the present embodiment, teachers' instruction video is preferably stored on public server, so that student and teacher look into
It askes.
In the present embodiment, a video clip includes at least a frame image.All are partitioned into from video clip
Raw single video clip is preferably but not limited to using manual segmentation or according to image middle school student's tables and chairs position distribution region
It is split, or processing is split using multiple target human region dividing method in existing video, publication number such as can be used
For method disclosed in CN108648198A.
In the present embodiment, face recognition algorithms use existing face recognition algorithms, it is preferred that are previously stored with class
All photos that should arrive student of one's parents, and with the associated personal information of each photo, by the facial image in single video clip
It is compared one by one with the photo prestored, if the similarity degree of the two is greater than or equal to 90%, i.e., it is believed that in the single video clip
Student be the associated student of the photo.Personal information is preferably but not limited to comprising student name and/or student number etc..
In the present embodiment, the classroom performance evaluation and set of each student's posture time of each student of output
In it is all do not meet classroom requirement the corresponding time tags of posture be storable in open server, so that related personnel looks into
It askes.Preferably, student's set Stu={ Stu is defined1,Stu2,Stu3,...,Stun, StukIndicate the classroom letter of k-th of student
Breath set, Stuk={ pk, class, [sk1,sk2,...,ski...,skN], [T1,T2,...,TN]};Class indicates course letter
Breath;pkIndicate the personal information of k-th of student;1≤k≤n;
It in the preferred embodiment of the present invention, in step s 2, further include the step that attendance is carried out to student attendance
Suddenly, it specifically includes:
Obtain all personal information that should arrive student in the classroom prestored, accumulative personal information and the institute that should each arrive student
The number for thering is single video clip to be consistent by student's personal information that recognition of face obtains, if the number is less than or equal to preset
Frequency threshold value, it is believed that this should arrive student's absence from duty and remind this that should attend class to student, if the number is greater than preset frequency threshold value, recognize
For this, should to arrive student attendance normal.
In the present embodiment, the attendance of automatic examination student participates in without teacher, alleviates teacher workload.
In the present embodiment, frequency threshold value is preferably but not limited to as 70%N-90%N.To the alerting pattern of student absent from duty
Server can be used and send reminder announced information from the smart machine of trend absence from duty student, remind when giving lessons of student's course
Between, location information.
In the preferred embodiment of the present invention, as shown in figure 3, in step s 2, being known based on gesture recognition model
Not Chu single video clip middle school student posture the step of specifically include:
Gesture recognition model is established, single video clip is inputted into gesture recognition model, gesture recognition model exports the list
The posture of people video clip middle school student;
The process for establishing gesture recognition model includes:
Step S21 constructs training dataset, is denoted as Vlabled;Training dataset VlabledPosture mark is provided with including multiple
The single video clip of label;
Step S22 extracts the video features that training data concentrates single video clip by video feature extraction module;
The video features of step S23, the single video clip concentrated with training data are input, with single video clip
Posture label is classification results, is trained and verifies to random forest grader, obtains gesture recognition model.
In the present embodiment, gesture recognition model is instructed using the random forest classification method based on deep learning
Practice, intelligence degree is high, participates in without artificial.
In the present embodiment, using random forest grader, the quantity of base decision tree is 10, all prediction roads of each tree
The length limit of diameter is 5, using this trained classifier come the posture for video of classifying.
In the preferred embodiment of the present invention, in the step s 21, the detailed process packet of training dataset is constructed
It includes:
Step S211 intercepts multiple video clips from existing student classroom video, is partitioned into each video clip
All single video clips are configured to single video clip collection, are denoted as V by the single video clip of all studentsunlabled;
Step S212 presets multiple postures, posture
S ∈ { reads, writes, pay attention to the class, stand up and answer a question, raise one's hand, say small words, play mobile phone, sleep };
Each single video clip is concentrated to be sent respectively to multiple interviewees single video clip, by interviewee to the list
People's video clip is given a mark with the degree that is consistent of each posture, is calculated each single video clip and is consistent with each posture degree score
Average value:
Wherein,Indicate that single video clip concentrates i-th ' a single video clip to be consistent with m-th of posture s (m) journey
Spend the average value of score;npIndicate interviewee's number to i-th ' a single video clip marking;I', m, j' are positive integer,
And 1≤m≤8,1≤j'≤np;
Posture label is arranged for single video clip in step S213:
If i-th ' a single video clip is consistent with m-th of posture s (m), the average value of degree score meets:
Posture label s then is set for i-th ' a single video clipi'And training dataset is added in i-th ' a single video clip
Vlabled, the posture label si'Are as follows:Wherein,For preset score threshold;
If i-th ' a single video clip is consistent with m-th of posture s (m), the average value of degree score is unsatisfactory for
Or i-th ' a single video clip be consistent with more than one posture degree score average value meetNot by i-th '
Training dataset V is added in a single video cliplabled。
In the present embodiment, multiple interviewees to each single video clip respectively with reading, write, pay attention to the class, stand up
It answers a question, raise one's hand, saying small words, playing mobile phone, this 8 postures of dozing off and be consistent degree marking, 5 points or 10 points of full marks can be used
System, full marks indicate that interviewee thinks in succession and are consistent completely that 0 point of expression interviewee thinks in succession not to be consistent completely.It is obtained to be preset
Dividing threshold value can be the 70% of full marks, when such as using 5 points of full marks,It can be 3.5 for preset score threshold.
In the present embodiment, using interviewee's marking system to data with existing collection carry out labeling, more hommization and
Accuracy.
In the preferred embodiment of the present invention, as shown in figure 3, video feature extraction module mentions in step S22
The process for taking training data to concentrate the video features of single video clip includes:
Step S221, single video clip a for i-th ' extract three of every frame image in i-th ' a single video clip
Posture is tieed up, dimensional posture set G is obtained;Preferably, by existing OpenPose method, (Github open source human body attitude is known
Other project) to estimate to obtain the dimensional posture in every frame image, dimensional posture schematic diagram is as shown in Figure 2.
G={ P1,P2,...,Pτ, τ is the totalframes that i-th ' a single video clip includes image, is positive integer;P1、
P2、…、PτRespectively indicate the 1st frame image in i-th ' a single video clip, the 2nd frame image ..., the three-dimensional appearance of τ frame image
Gesture;
Step S222 obtains the further feature F of dimensional posture set G by existing shot and long term memory modelsdeep;
Step S223 extracts the feature F that attends class from dimensional posture set Gclass, feature of attending class FclassIncluding posture feature
FposeWith motion feature Fmove;
Posture featureWherein, Ft,poseIndicate the posture feature of t-th of picture frame, Ft,pose=
{f1,f2,...,f16, in t-th of picture frame, f1Indicate that human body occupies the size of picture, i.e. human body in t-th of picture frame
Area pixel point is total and/or human region accounts for the ratio of whole image frame;f2Indicate left shoulder c to neck b line and right shoulder q to neck
The angle that portion's b line is formed, as shown in Fig. 2, being angle ∠ cbq;f3Indicate left shoulder c to neck b line and head a to neck b
The angle that line is formed, as shown in Fig. 2, being angle ∠ cba;f4Indicate right shoulder q to neck b line and head a to neck b line
The angle of formation, as shown in Fig. 2, being angle ∠ qba;f5Indicate that head a is formed with back d to neck b line to neck b line
Angle, as shown in Fig. 2, be angle ∠ abd;f6Indicate neck b to back d line and coccyx root f to back d line shape
At angle, as shown in Fig. 2, be angle ∠ bdf;f7Indicate that left shoulder c is formed with left hand g to left elbow e line to left elbow e line
Angle, as shown in Fig. 2, be angle ∠ ceg;f8Indicate the angle that right shoulder q is formed to right elbow p line and right hand o to right elbow p line
Degree, as shown in Fig. 2, being angle ∠ qpo; f9Indicate the angle that the portion left stern h is formed to left knee u line and left foot to left knee u line
Degree, as shown in Fig. 2, being angle ∠ huw;f10Indicate the angle that right hips are formed to right knee m line and right crus of diaphragm v to right knee m line,
As shown in Fig. 2, being angle ∠ zmv;f11Right hand o is indicated at a distance from the f of coccyx root, as shown in Fig. 2, being distance of;f12It indicates
Left hand g is at a distance from the f of coccyx root, as shown in Fig. 2, being distance hf;f13Right crus of diaphragm v is at a distance from the f of coccyx root for expression, such as Fig. 2
It is shown, it is distance vf;f14Left foot w is indicated at a distance from the f of coccyx root, as shown in Fig. 2, being distance wf;f15Indicate two hands with
The area for the triangle that neck b is surrounded, as shown in Fig. 2, being the area of triangle Δ obg;f16Indicate two feet and coccyx root
The triangle area that portion f is surrounded is as shown in Fig. 2, be the area of triangle Δ vfw;1≤t≤τ;
Motion featureWherein, Ft,moveIndicate the motion feature of t-th of picture frame, Ft,move=
{f17,f18,...,f25, in t-th of picture frame, f17Indicate right hand o speed, f18Indicate right hand o acceleration, f19Indicate the right hand
O acts rapid degree, f20Indicate left hand g speed, f21Indicate left hand g acceleration, f22Indicate that left hand g acts rapid degree, f23
Indicate head a speed, f24Indicate head a acceleration, f25Indicate that head a acts rapid degree;
Single video clip viThe feature F that attends classclass=Fpose+Fmove;
Step S224, single video clip viVideo features are as follows: Ftotal=Fdeep+Fclass。
In the present embodiment, for two hands and these three joints head a, these three joints in each frame image are recorded
Position, then single video clip just has the displacement record changed over time in these three joints, and displacement record can use position
Set vector expression.To the position vector first derivation in a certain joint, obtain be the joint speed, to the position in a certain joint to
Second order derivation is measured, what is obtained is the acceleration of the joint motions;Three rank derivation of position vector to a certain joint, what is obtained is to add
The size degree of velocity variations, referred to as acts rapid degree.
In the present embodiment, sought from the dimensional posture of every frame image it is multiple come posture feature of building together so as to
The identification of raw posture is more acurrate and scientific.Certainly, due to student's sitting, the movements such as move, fall and influence, in one frame of image may be used
F can be sought completely1To f25Totally 25 characteristic parameters, the characteristic parameter that cannot seek obtaining at this time can be by before seeking
The average value of this feature parameter (multiple historical record values of this feature parameter) obtains in multiple image.
In the preferred embodiment of the present invention, in step s3, the posture time set based on all students obtains
The classroom for obtaining each student shows the step of evaluating and specifically includes:
Step S31, if identifying the posture s of the single video clip middle school student based on gesture recognition model are as follows:
Step S32 seeks listen to the teacher state of each student on classroom, classroom interaction degree, classroom liveness and not specially
Heart degree;
The state L that listens to the teacher of k-th of studentkAre as follows:Wherein, the LkiFor k-th of student i-th one
Video clip viIn state of listening to the teacher,skiIndicate k-th of student in i-th of single piece of video
Section viIn posture;
Preferably, if k-th of student is in i-th of single video clip viIn the state of listening to the teacher can not identify, then by
Raw history is attended class listen to the teacher state of the status information as current single video clip, such as by the single piece of video of (i-1)-th or i-2
Listen to the teacher state of the state as current single video clip of listening to the teacher in section.
Classroom interaction degree is by raising one's hand and standing up to answer a question and determine in posture of attending class, the classroom interaction of k-th of student
Degree TkAre as follows:Wherein,For preset first weight parameter;Ωhand_kIt indicates k-th
Student's posture in N number of video clip is the accounting raised one's hand,Ωstand_kIndicate k-th of student
Posture is the accounting answered a question of standing up in N number of video clip,
The classroom liveness B of k-th of studentkAre as follows: Bk=max (0, min (1,1- Δ Bk));Wherein, Δ BkIt indicates k-th
The classroom liveness B of studentkVariable quantity, Δ Bk=δ1*(τread_k+τwrite_k+τlisten_k)+δ2*(τtalk_k+τphone_k+
τsleep_k)-δ3*τhand_k-δ4*τstand_k, δ1Indicate the preset first coefficient of rewards and punishment factor, δ2Indicate preset second rewards and punishments system
The number factor, δ3Indicate the preset third coefficient of rewards and punishment factor, δ4Indicate the preset 4th coefficient of rewards and punishment factor, τread_k、τwrite_k、
τlisten_k、τtalk_k、τphone_k、τsleep_k、τhand_k、τstand_kRespectively indicate k-th of student continuous N in N number of video clip
Video clip detecting state is reading, writes, pay attention to the class, says small words, play mobile phone, sleep, raise one's hand, the state answered a question of standing up
Number, 1≤M≤N;Preferably, M is 8 times to 15 times, can be 10 times.
Student classroom liveness B by student whether for a long time keep reads, write, pay attention to the class, say it is small words, object for appreciation mobile phone,
It the posture of doze and raises one's hand, stand up and answer a question to determine.In classroom, the high student of liveness tends not to read a book always
Or pay attention to the class, but read a book and take notes while paying attention to the class.The a reference value of B is 1, and range is between 0 to 1.
The case where persistently saying small words according to above-mentioned student, play mobile phone, sleep defines the not attentive concentration of student, k-th
Raw not attentive concentration ZkAre as follows: Zk=τtalk_k+τphone_k+τsleep_k;
Step S33, based on listen to the teacher state of all students on classroom, classroom interaction degree, classroom liveness and not specially
Heart degree obtains the classroom performance evaluation of each student, specifically includes:
Step S331 constructs initial decision Matrix C;
N is student's total number of persons;
Initial decision Matrix C is normalized, acquisition specified decision Matrix C ';
Weight matrix W is arranged in step S332:
Wherein, η1Indicate the weight of state of listening to the teacher, η2Indicate the power of classroom interaction degree
Weight, η3Indicate the weight of classroom liveness, η4Indicate the weight of not attentive concentration;
Step S333 calculates weighted decision matrix D,
Solve the positive ideal solution of weighted decision matrix DAnd minus ideal result
Wherein, the columns serial number of j expression weighted decision matrix D, j=1,2,3,4;
Step S334 calculates the Euclidean distance between each student and positive and negative ideal value;
The Euclidean distance of k-th student and positive ideal solutionAre as follows:
The Euclidean distance of k-th of student and minus ideal resultAre as follows:
The classroom performance of k-th of student is evaluated as Vk:
In the present embodiment, from the state of listening to the teacher, classroom interaction degree, classroom liveness and the not dimension of attentive concentration this four
Degree goes the performance of evaluation student at school, goes to evaluate according to multiple dimensions, can be directly obtained by way of decision matrix every
A student is a kind of more scientific and reasonable evaluation method closest to the degree of fitst water student.According to the classroom table of student
A kind of scoring is now designed, teacher is helped to record the classroom performance of student and is scientifically scored, is conducive to improve religion
Learn quality.
In the preferred embodiment of the present invention, in step s 4, the posture for not meeting classroom requirement is small including saying
Words play mobile phone and sleep.
In the present embodiment, it is preferred that further include:
If identifying that the posture of single video clip middle school student does not meet classroom and requires based on gesture recognition model, to the list
Student in people's video clip reminds, further, if the student in the single video clip is in M1 before single views
Posture in frequency segment does not meet classroom requirement, and teacher is reminded to remind the student, and M1 is preset positive integer, 1≤M1≤N.
M1 is preferably but not limited to 5-10.
The invention also discloses a kind of student classroom state evaluation systems, in a preferred embodiment, such as Fig. 4 and figure
Shown in 5, which includes the first camera 1 for shooting all students on classroom, shoots second of teachers' instruction content on classroom
Camera 2 and server 3;
The teachers' instruction view that server 3 receives the student classroom video of the first camera 1 output and second camera 2 exports
Frequently, the classroom performance evaluation of each student and according to above-mentioned student's method for evaluating state based on gesture recognition is obtained, and is saved
The classroom that do not meet of student requires the corresponding time tag of posture and teachers' instruction video.
In the present embodiment, it is preferred that the first camera 1 is located in front of classroom and faces all students, the second camera shooting
First 2 are located at classroom rear and face teacher, can take blackboard or projector.Server 3 can be located at classroom or school control
Center, with the first camera 1 and the wired or wireless connection of second camera 2.
It in the preferred embodiment of the present invention, further include multiple students end that connection communication is established with server 3
End equipment 4 and teacher's terminal device.
In the present embodiment, student terminal equipment 4 and teacher's terminal device and server 3 are wirelessly connected.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms are not
Centainly refer to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be any
One or more embodiment or examples in can be combined in any suitable manner.
Although an embodiment of the present invention has been shown and described, it will be understood by those skilled in the art that: not
A variety of change, modification, replacement and modification can be carried out to these embodiments in the case where being detached from the principle of the present invention and objective, this
The range of invention is defined by the claims and their equivalents.
Claims (10)
1. a kind of student's method for evaluating state based on gesture recognition characterized by comprising
Step S1 obtains the classroom video and teachers' instruction video of all students on classroom in real time, stores teachers' instruction video;
Compartment of terrain intercepts multiple video clips from the classroom video;
Each video clip is associated with the time tag for recording the video clip actual photographed time;
Each video clip is handled as follows in step S2:
It is partitioned into the single video clip of all students in video clip;
For any single video clip, identify that the personal of the single video clip middle school student is believed using face recognition algorithms
Breath, the posture of the single video clip middle school student is identified based on gesture recognition model, by the personal information and posture phase
Association;
Step S3 utilizes the personal information obtained from all video clips and posture structure associated with the personal information
The posture time set for building each student, if the posture time collection of k-th of student is combined into Uk={ [sk1,sk2,...,ski...,
skN], [T1,T2,...,Ti,...,TN]};
Wherein, 1≤k≤n, n indicate the total number of persons of student, are positive integer;N indicates the total quantity of video clip, is positive integer;ski
Indicate posture of k-th of student in i-th of video clip;TiIndicate the time tag of i-th of video clip, 1≤i≤N;ski
With TiIt corresponds;
The classroom that posture time set based on all students obtains each student shows evaluation;
Step S4 is exported in the classroom performance evaluation and set of each student's posture time of each student and all is not met class
The corresponding time tag of posture that hall requires.
2. student's method for evaluating state based on gesture recognition as described in claim 1, which is characterized in that in the step S2
In, further include the steps that carrying out attendance to student attendance, specifically include:
Obtain all personal information that should arrive student in the classroom prestored, the accumulative personal information that should each arrive student and all lists
The number that people's video clip is consistent by student's personal information that recognition of face obtains, if the number is less than or equal to preset time
Number threshold value, it is believed that this should arrive student's absence from duty and remind this that should attend class to student, if the number is greater than preset frequency threshold value, recognize
For this, should to arrive student attendance normal.
3. student's method for evaluating state based on gesture recognition as described in claim 1, which is characterized in that in the step S2
In, the step of identifying the posture of the single video clip middle school student based on gesture recognition model, specifically includes:
Establish gesture recognition model, by single video clip input gesture recognition model, gesture recognition model output it is described one
The posture of video clip middle school student;
The process for establishing gesture recognition model includes:
Step S21 constructs training dataset, is denoted as Vlabled;The training dataset VlabledPosture mark is provided with including multiple
The single video clip of label;
Step S22 extracts the video features that training data concentrates single video clip by video feature extraction module;Step
S23, the video features for the single video clip concentrated with training data are input, are point with the posture label of single video clip
Class obtains gesture recognition model as a result, be trained and verify to random forest grader.
4. student's method for evaluating state based on gesture recognition as claimed in claim 3, which is characterized in that in the step
In S21, the detailed process for constructing training dataset includes:
Step S211 intercepts multiple video clips from existing student classroom video, is partitioned into each video clip and owns
All single video clips are configured to single video clip collection, are denoted as V by the single video clip of studentunlabled;
Step S212 presets multiple postures, posture
S ∈ { reads, writes, pay attention to the class, stand up and answer a question, raise one's hand, say small words, play mobile phone, sleep };
Concentrate each single video clip to be sent respectively to multiple interviewees single video clip, by interviewee to it is described one
Video clip is given a mark with the degree that is consistent of each posture, is calculated each single video clip and is consistent with each posture degree score
Average value:
Wherein, ri' s(m)Indicate that single video clip concentrates i-th ' a single video clip degree that is consistent with m-th of posture s (m) to obtain
The average value divided;npIndicate interviewee's number to i-th ' a single video clip marking;I', m, j' are positive integer, and 1≤
M≤8,1≤j'≤np;
Posture label is arranged for single video clip in step S213:
If i-th ' a single video clip is consistent with m-th of posture s (m), the average value of degree score meets:Then it is
Posture label s is arranged in i-th ' a single video clipi'And training dataset V is added in i-th ' a single video cliplabled, institute
State posture label si'Are as follows:Wherein,For preset score threshold;
If i-th ' a single video clip is consistent with m-th of posture s (m), the average value of degree score is unsatisfactory forOr
I-th ' a single video clip be consistent with more than one posture degree score average value meetNot by i-th ' a list
Training dataset V is added in people's video cliplabled。
5. student's method for evaluating state based on gesture recognition as claimed in claim 3, which is characterized in that in the step
In S22, video feature extraction module extracts training data and concentrates the processes of the video features of single video clip to include:
Step S221, single video clip a for i-th ' extract the three-dimensional appearance of every frame image in i-th ' a single video clip
Gesture obtains dimensional posture set G;
G={ the P1,P2,...,Pτ, τ is the totalframes that i-th ' a single video clip includes image, is positive integer;P1、
P2、…、PτRespectively indicate the 1st frame image in i-th ' a single video clip, the 2nd frame image ..., the three-dimensional appearance of τ frame image
Gesture;
Step S222 obtains the further feature F of dimensional posture set G by shot and long term memory modelsdeep;
Step S223 extracts the feature F that attends class from dimensional posture set Gclass, the feature F that attends classclassIncluding posture feature
FposeWith motion feature Fmove;
The posture featureWherein, Ft,poseIndicate the posture feature of t-th of picture frame, Ft,pose=
{f1,f2,...,f16, in t-th of picture frame, f1Indicate that human body occupies the size of picture, f in t-th of picture frame2It indicates
The angle that left shoulder is formed to neck line and right shoulder to neck line, f3Indicate left shoulder to neck line and head to neck line
The angle of formation, f4Indicate the angle that right shoulder is formed to neck line and head to neck line, f5Indicate head to neck line
The angle formed with back to neck line, f6Indicate the angle that neck is formed to back line and coccyx root to back line
Degree, f7Indicate the angle that left shoulder is formed to left elbow line and left hand to left elbow line, f8Indicate that right shoulder is arrived to right elbow line with the right hand
The angle that right elbow line is formed, f9Indicate the angle that left buttocks is formed to left knee line and left foot to left knee line, f10Indicate right
The angle that buttocks is formed to right knee line and right crus of diaphragm to right knee line, f11The expression right hand is at a distance from coccyx root, f12Indicate left
Hand is at a distance from coccyx root, f13Expression right crus of diaphragm is at a distance from coccyx root, f14Expression left foot is at a distance from coccyx root, f15
Indicate the area for the triangle that two hands and neck are surrounded, f16Indicate the gore that two feet and coccyx root are surrounded
Product;1≤t≤τ;
The motion featureWherein, Ft,moveIndicate the motion feature of t-th of picture frame, Ft,move=
{f17,f18,...,f25, in t-th of picture frame, f17Indicate right hand speed, f18Indicate right hand acceleration, f19Indicate that the right hand is dynamic
Make rapid degree, f20Indicate left hand speed, f21Indicate left hand acceleration, f22Indicate that left hand acts rapid degree, f23Indicate head
Speed, f24Indicate head acceleration, f25Indicate the rapid degree of headwork;
Single video clip viThe feature F that attends classclass=Fpose+Fmove;
Step S224, single video clip viVideo features are as follows: Ftotal=Fdeep+Fclass。
6. student's method for evaluating state based on gesture recognition as described in claim 1, which is characterized in that in the step S3
In, the classroom that the posture time set based on all students obtains each student shows the step of evaluating and specifically includes:
Step S31, if identifying the posture s of the single video clip middle school student based on gesture recognition model are as follows:
Step S32 seeks listen to the teacher state, classroom interaction degree, classroom liveness and inwholwe-hearted journey of each student on classroom
Degree;
The state L that listens to the teacher of k-th of studentkAre as follows:Wherein, the LkiIt is k-th of student in i-th of single video
Segment viIn state of listening to the teacher,skiIndicate k-th of student in i-th of single video clip vi
In posture;
The classroom interaction degree T of k-th of studentkAre as follows:Wherein,For preset first power
Weight parameter;Ωhand_kIndicate that k-th of student posture in N number of video clip is the accounting raised one's hand,Ωstand_kIndicate that k-th of student posture in N number of video clip is the accounting for of answering a question of standing up
Than,
The classroom liveness B of k-th of studentkAre as follows: Bk=max (0, min (1,1- Δ Bk));Wherein, Δ BkIndicate k-th of student's
Classroom liveness BkVariable quantity, Δ Bk=δ1*(τread_k+τwrite_k+τlisten_k)+δ2*(τtalk_k+τphone_k+τsleep_k)-
δ3*τhand_k-δ4*τstand_k, δ1Indicate the preset first coefficient of rewards and punishment factor, δ2Indicate the preset second coefficient of rewards and punishment factor, δ3
Indicate the preset third coefficient of rewards and punishment factor, δ4Indicate the preset 4th coefficient of rewards and punishment factor, τread_k、τwrite_k、τlisten_k、
τtalk_k、τphone_k、τsleep_k、τhand_k、τstand_kRespectively indicate k-th of student continuous N piece of video in N number of video clip
Section detecting state is reading, writes, pays attention to the class, say small words, play mobile phone, sleep, raise one's hand, time for the state answered a question of standing up
Number, 1≤M≤N;
The not attentive concentration Z of k-th of studentkAre as follows: Zk=τtalk_k+τphone_k+τsleep_k;
Step S33, state of listening to the teacher, classroom interaction degree, classroom liveness and inwholwe-hearted journey based on all students on classroom
The classroom that degree obtains each student shows evaluation, specifically includes:
Step S331 constructs initial decision Matrix C;
It is describedN is student's total number of persons;
Initial decision Matrix C is normalized, acquisition specified decision Matrix C ';
It is described
Weight matrix W is arranged in step S332:
It is describedWherein, η1Indicate the weight of state of listening to the teacher, η2Indicate the power of classroom interaction degree
Weight, η3Indicate the weight of classroom liveness, η4Indicate the weight of not attentive concentration;
Step S333 calculates weighted decision matrix D,
It is described
Solve the positive ideal solution of weighted decision matrix DAnd minus ideal result
It is describedIt is described
Wherein, the columns serial number of j expression weighted decision matrix D, j=1,2,3,4;
Step S334 calculates the Euclidean distance between each student and positive and negative ideal value;
The Euclidean distance of k-th student and positive ideal solutionAre as follows:
The Euclidean distance of k-th of student and minus ideal resultAre as follows:
The classroom performance of k-th of student is evaluated as Vk:
7. student's method for evaluating state based on gesture recognition as described in claim 1, which is characterized in that in the step S4
In, the posture for not meeting classroom requirement includes saying small words, playing mobile phone and sleep.
8. student's method for evaluating state based on gesture recognition as claimed in claim 7, which is characterized in that further include:
If identifying that the posture of the single video clip middle school student does not meet classroom and requires based on gesture recognition model, to described
Student in single video clip reminds, further, if the student in the single video clip is single at M1 before
Posture in people's video clip does not meet classroom requirement, and teacher is reminded to remind the student, and M1 is preset positive integer, 1≤M1
≤N。
9. a kind of student classroom state evaluation system, which is characterized in that the first camera including all students on shooting classroom,
Shoot the second camera and server of teachers' instruction content on classroom;
The server receives the student classroom video of the first camera output and the teachers' instruction video of second camera output,
And the classroom table of each student is obtained according to student's method for evaluating state described in one of claim 1-8 based on gesture recognition
It now evaluates, and the classroom that do not meet for saving student requires the corresponding time tag of posture and teachers' instruction video.
10. student classroom state evaluation system as claimed in claim 9, which is characterized in that further include establishing to connect with server
Connect the multiple student terminal equipment and teacher's terminal device of letter.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910758393.5A CN110443226B (en) | 2019-08-16 | 2019-08-16 | Student state evaluation method and system based on posture recognition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910758393.5A CN110443226B (en) | 2019-08-16 | 2019-08-16 | Student state evaluation method and system based on posture recognition |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110443226A true CN110443226A (en) | 2019-11-12 |
CN110443226B CN110443226B (en) | 2022-01-25 |
Family
ID=68435975
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910758393.5A Active CN110443226B (en) | 2019-08-16 | 2019-08-16 | Student state evaluation method and system based on posture recognition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110443226B (en) |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111062844A (en) * | 2020-03-17 | 2020-04-24 | 浙江正元智慧科技股份有限公司 | Intelligent control system for smart campus |
CN111144255A (en) * | 2019-12-18 | 2020-05-12 | 华中科技大学鄂州工业技术研究院 | Method and device for analyzing non-language behaviors of teacher |
CN111507873A (en) * | 2020-04-14 | 2020-08-07 | 四川聚阳科技集团有限公司 | Classroom participation degree evaluation method based on sound and image joint sampling |
CN111507227A (en) * | 2020-04-10 | 2020-08-07 | 南京汉韬科技有限公司 | Multi-student individual segmentation and state autonomous identification method based on deep learning |
CN111553218A (en) * | 2020-04-20 | 2020-08-18 | 南京医科大学 | Intelligent medical skill teaching monitoring system based on human body posture recognition |
CN111611854A (en) * | 2020-04-16 | 2020-09-01 | 杭州电子科技大学 | Classroom condition evaluation method based on pattern recognition |
CN111680558A (en) * | 2020-04-29 | 2020-09-18 | 北京易华录信息技术股份有限公司 | Learning special attention assessment method and device based on video images |
CN111738177A (en) * | 2020-06-28 | 2020-10-02 | 四川大学 | Student classroom behavior identification method based on attitude information extraction |
CN112686154A (en) * | 2020-12-29 | 2021-04-20 | 杭州晨安科技股份有限公司 | Student standing detection method based on head detection and picture sequence |
CN112862643A (en) * | 2021-03-01 | 2021-05-28 | 深圳市微幼科技有限公司 | Multimedia remote education platform system |
CN112905811A (en) * | 2021-02-07 | 2021-06-04 | 广州录臻科技有限公司 | Teaching audio and video pushing method and system based on student classroom behavior analysis |
CN112990105A (en) * | 2021-04-19 | 2021-06-18 | 北京优幕科技有限责任公司 | Method and device for evaluating user, electronic equipment and storage medium |
CN114285966A (en) * | 2021-12-07 | 2022-04-05 | 浙江东隆教育科技有限公司 | Labor education related monitoring data processing method and system |
CN114440884A (en) * | 2022-04-11 | 2022-05-06 | 天津果实科技有限公司 | Intelligent analysis method for human body posture for intelligent posture correction equipment |
CN116055684A (en) * | 2023-01-18 | 2023-05-02 | 訸和文化科技(苏州)有限公司 | Online physical education system based on picture monitoring |
CN116757524A (en) * | 2023-05-08 | 2023-09-15 | 广东保伦电子股份有限公司 | Teacher teaching quality evaluation method and device |
CN117196237A (en) * | 2023-09-19 | 2023-12-08 | 武汉盟游网络科技有限公司 | Information management method and device based on cloud computing |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120052476A1 (en) * | 2010-08-27 | 2012-03-01 | Arthur Carl Graesser | Affect-sensitive intelligent tutoring system |
US20150302866A1 (en) * | 2012-10-16 | 2015-10-22 | Tal SOBOL SHIKLER | Speech affect analyzing and training |
CN107085721A (en) * | 2017-06-26 | 2017-08-22 | 厦门劢联科技有限公司 | A kind of intelligence based on Identification of Images patrols class management system |
CN107958351A (en) * | 2017-12-26 | 2018-04-24 | 重庆大争科技有限公司 | Teaching quality assessment cloud service platform |
CN108564022A (en) * | 2018-04-10 | 2018-09-21 | 深圳市唯特视科技有限公司 | A kind of more personage's pose detection methods based on positioning classification Recurrent networks |
CN108875606A (en) * | 2018-06-01 | 2018-11-23 | 重庆大学 | A kind of classroom teaching appraisal method and system based on Expression Recognition |
CN109035089A (en) * | 2018-07-25 | 2018-12-18 | 重庆科技学院 | A kind of Online class atmosphere assessment system and method |
CN109359539A (en) * | 2018-09-17 | 2019-02-19 | 中国科学院深圳先进技术研究院 | Attention appraisal procedure, device, terminal device and computer readable storage medium |
CN109685692A (en) * | 2019-01-15 | 2019-04-26 | 山东仁博信息科技有限公司 | A kind of noninductive acquisition and analysis system of various dimensions student learning behavior |
-
2019
- 2019-08-16 CN CN201910758393.5A patent/CN110443226B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120052476A1 (en) * | 2010-08-27 | 2012-03-01 | Arthur Carl Graesser | Affect-sensitive intelligent tutoring system |
US20150302866A1 (en) * | 2012-10-16 | 2015-10-22 | Tal SOBOL SHIKLER | Speech affect analyzing and training |
CN107085721A (en) * | 2017-06-26 | 2017-08-22 | 厦门劢联科技有限公司 | A kind of intelligence based on Identification of Images patrols class management system |
CN107958351A (en) * | 2017-12-26 | 2018-04-24 | 重庆大争科技有限公司 | Teaching quality assessment cloud service platform |
CN108564022A (en) * | 2018-04-10 | 2018-09-21 | 深圳市唯特视科技有限公司 | A kind of more personage's pose detection methods based on positioning classification Recurrent networks |
CN108875606A (en) * | 2018-06-01 | 2018-11-23 | 重庆大学 | A kind of classroom teaching appraisal method and system based on Expression Recognition |
CN109035089A (en) * | 2018-07-25 | 2018-12-18 | 重庆科技学院 | A kind of Online class atmosphere assessment system and method |
CN109359539A (en) * | 2018-09-17 | 2019-02-19 | 中国科学院深圳先进技术研究院 | Attention appraisal procedure, device, terminal device and computer readable storage medium |
CN109685692A (en) * | 2019-01-15 | 2019-04-26 | 山东仁博信息科技有限公司 | A kind of noninductive acquisition and analysis system of various dimensions student learning behavior |
Non-Patent Citations (6)
Title |
---|
JANEZ ZALETELJ: "Estimation of Students’ Attention in the Classroom From Kinect Features", 《10TH INTERNATIONAL SYMPOSIUM ON IMAGE AND SIGNAL PROCESSING AND ANALYSIS (ISPA 2017)》 * |
S L HAPPY等: "Automated Alertness and Emotion Detection for Empathic Feedback During E-Learning", 《2013 IEEE FIFTH INTERNATIONAL CONFERENCE ON TECHNOLOGY FOR EDUCATION》 * |
李鑫鑫: "基于视频的人体行为分析", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
杨露菁等编著: "《智能图像处理及应用》", 31 March 2019, 中国铁道出版社 * |
陈晋音等: "基于深度学习的智能教学系统的设计与研究", 《计算机科学》 * |
骆祖莹等: "课堂教学自动评价及其初步研究成果", 《现代教育技术》 * |
Cited By (26)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111144255B (en) * | 2019-12-18 | 2024-04-19 | 华中科技大学鄂州工业技术研究院 | Analysis method and device for non-language behaviors of teacher |
CN111144255A (en) * | 2019-12-18 | 2020-05-12 | 华中科技大学鄂州工业技术研究院 | Method and device for analyzing non-language behaviors of teacher |
CN111062844A (en) * | 2020-03-17 | 2020-04-24 | 浙江正元智慧科技股份有限公司 | Intelligent control system for smart campus |
CN111507227A (en) * | 2020-04-10 | 2020-08-07 | 南京汉韬科技有限公司 | Multi-student individual segmentation and state autonomous identification method based on deep learning |
CN111507227B (en) * | 2020-04-10 | 2023-04-18 | 南京汉韬科技有限公司 | Multi-student individual segmentation and state autonomous identification method based on deep learning |
CN111507873A (en) * | 2020-04-14 | 2020-08-07 | 四川聚阳科技集团有限公司 | Classroom participation degree evaluation method based on sound and image joint sampling |
CN111611854A (en) * | 2020-04-16 | 2020-09-01 | 杭州电子科技大学 | Classroom condition evaluation method based on pattern recognition |
CN111611854B (en) * | 2020-04-16 | 2023-09-01 | 杭州电子科技大学 | Classroom condition evaluation method based on pattern recognition |
CN111553218A (en) * | 2020-04-20 | 2020-08-18 | 南京医科大学 | Intelligent medical skill teaching monitoring system based on human body posture recognition |
CN111680558A (en) * | 2020-04-29 | 2020-09-18 | 北京易华录信息技术股份有限公司 | Learning special attention assessment method and device based on video images |
CN111738177A (en) * | 2020-06-28 | 2020-10-02 | 四川大学 | Student classroom behavior identification method based on attitude information extraction |
CN111738177B (en) * | 2020-06-28 | 2022-08-02 | 四川大学 | Student classroom behavior identification method based on attitude information extraction |
CN112686154A (en) * | 2020-12-29 | 2021-04-20 | 杭州晨安科技股份有限公司 | Student standing detection method based on head detection and picture sequence |
CN112686154B (en) * | 2020-12-29 | 2023-03-07 | 杭州晨安科技股份有限公司 | Student standing detection method based on head detection and picture sequence |
CN112905811A (en) * | 2021-02-07 | 2021-06-04 | 广州录臻科技有限公司 | Teaching audio and video pushing method and system based on student classroom behavior analysis |
CN112862643A (en) * | 2021-03-01 | 2021-05-28 | 深圳市微幼科技有限公司 | Multimedia remote education platform system |
CN112990105B (en) * | 2021-04-19 | 2021-09-21 | 北京优幕科技有限责任公司 | Method and device for evaluating user, electronic equipment and storage medium |
CN112990105A (en) * | 2021-04-19 | 2021-06-18 | 北京优幕科技有限责任公司 | Method and device for evaluating user, electronic equipment and storage medium |
CN114285966A (en) * | 2021-12-07 | 2022-04-05 | 浙江东隆教育科技有限公司 | Labor education related monitoring data processing method and system |
CN114285966B (en) * | 2021-12-07 | 2024-03-29 | 浙江东隆教育科技有限公司 | Method and system for processing monitoring data related to labor education |
CN114440884A (en) * | 2022-04-11 | 2022-05-06 | 天津果实科技有限公司 | Intelligent analysis method for human body posture for intelligent posture correction equipment |
CN116055684A (en) * | 2023-01-18 | 2023-05-02 | 訸和文化科技(苏州)有限公司 | Online physical education system based on picture monitoring |
CN116055684B (en) * | 2023-01-18 | 2023-12-12 | 广州乐体科技有限公司 | Online physical education system based on picture monitoring |
CN116757524A (en) * | 2023-05-08 | 2023-09-15 | 广东保伦电子股份有限公司 | Teacher teaching quality evaluation method and device |
CN116757524B (en) * | 2023-05-08 | 2024-02-06 | 广东保伦电子股份有限公司 | Teacher teaching quality evaluation method and device |
CN117196237A (en) * | 2023-09-19 | 2023-12-08 | 武汉盟游网络科技有限公司 | Information management method and device based on cloud computing |
Also Published As
Publication number | Publication date |
---|---|
CN110443226B (en) | 2022-01-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110443226A (en) | A kind of student's method for evaluating state and system based on gesture recognition | |
CN106228982B (en) | A kind of interactive learning system and exchange method based on education services robot | |
CN110991381B (en) | Real-time classroom student status analysis and indication reminding system and method based on behavior and voice intelligent recognition | |
CN107030691B (en) | Data processing method and device for nursing robot | |
CN108875606A (en) | A kind of classroom teaching appraisal method and system based on Expression Recognition | |
Breazeal et al. | Challenges in building robots that imitate people | |
US10957325B2 (en) | Method and apparatus for speech interaction with children | |
WO2018121624A1 (en) | Robot, server and man-machine interaction method | |
Kanda et al. | A two-month field trial in an elementary school for long-term human–robot interaction | |
CN108399376A (en) | Student classroom learning interest intelligent analysis method and system | |
CN109637207A (en) | A kind of preschool education interactive teaching device and teaching method | |
CN111754368A (en) | College teaching evaluation method and college teaching evaluation system based on edge intelligence | |
Nguyen et al. | Online feedback system for public speakers | |
US11160717B2 (en) | Device and method for instilling intrinsic motivation regarding eye contact in children affected by eye contact disorders | |
CN102945624A (en) | Intelligent video teaching system based on cloud calculation model and expression information feedback | |
JP7106159B2 (en) | Image education system and learning support method using artificial intelligence technology | |
CN111291613B (en) | Classroom performance evaluation method and system | |
WO2019168162A1 (en) | Information processing device, information processing method, and program | |
McColl et al. | Classifying a person’s degree of accessibility from natural body language during social human–robot interactions | |
He et al. | An educational robot system of visual question answering for preschoolers | |
JP2016100033A (en) | Reproduction control apparatus | |
Ma et al. | A deep learning approach for online learning emotion recognition | |
US20180277002A1 (en) | Reading and Contingent Response Educational and Entertainment Method and Apparatus | |
CN110766989A (en) | Ancient poetry teaching system based on artificial intelligence | |
CN111178263A (en) | Real-time expression analysis method and device |
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 |