CN112101231A

CN112101231A - Learning behavior monitoring method, terminal, small program and server

Info

Publication number: CN112101231A
Application number: CN202010974828.2A
Authority: CN
Inventors: 李浩文
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2020-09-16
Filing date: 2020-09-16
Publication date: 2020-12-18

Abstract

The embodiment of the application provides a learning behavior monitoring method, a terminal, an applet and a server, and relates to the field of online education. The method comprises the following steps: displaying the unique terminal identification, and establishing a binding relationship between the applet and the account information of the applet according to the unique terminal identification; and acquiring the target learning duration set by the small program with the binding relationship, and collecting the learning information of the students by combining the target learning duration. The embodiment of the application utilizes convenience and terminal linkage interaction of the applet, helps parents to supervise and manage the learning condition of students, helps to improve the work completion efficiency of students, develops a good time management idea, and helps parents to know the learning condition of children.

Description

Learning behavior monitoring method, terminal, small program and server

Technical Field

The application relates to the technical field of online education, in particular to a learning behavior monitoring method, a terminal, an applet and a server.

Background

Cloud Computing reduction (CCEDU) refers to an Education platform service based on Cloud Computing business model application. On the cloud platform, all education institutions, training institutions, enrollment service institutions, propaganda institutions, industry associations, management institutions, industry media, legal structures and the like are integrated into a resource pool in a centralized cloud mode, all resources are mutually displayed and interacted and communicated according to needs to achieve intentions, so that education cost is reduced, and efficiency is improved.

With the widespread use of transmission media such as televisions, computers, and the internet, many online learning methods for cloud education via a communication network have been developed. The online learning can break through the time and space limitations, so that students can learn courses at any time and any place.

The existing online learning mainly provides homework correction for a teacher end or evaluates and counts the learning condition of students, and is less provided for a parent to use, and actually, the parent pays more attention to the condition of doing homework for the students than the teacher, and at present, the parent can only pass through a mode of in-person supervision when wanting to stipulate the learning time for the students, and the mode has the problems of low efficiency and large parent burden.

Disclosure of Invention

Embodiments of the present invention provide a monitoring method, terminal, applet and server for learning behaviour that overcome the above problems or at least partially address the above problems.

In a first aspect, a method for monitoring learning behavior is provided, which is applied to a terminal, and includes:

the method comprises the steps that a unique identification of a terminal is displayed, and the applet establishes a binding relationship with account information of the applet according to the unique identification of the terminal;

and acquiring the target learning duration set by the small program with the binding relationship, and acquiring the learning information of the student according to the target learning duration.

Further, acquiring the target learning period set by the applet includes:

receiving a target learning duration pushed by an applet having a binding relationship.

Further, acquiring the target learning period set by the applet includes:

sending a time length obtaining request to a server, wherein the time length obtaining request comprises a unique identifier of a terminal, and the server returns a target learning time length corresponding to the unique identifier of the terminal to the terminal according to the time length obtaining request;

the server stores the binding relationship between the account information of the applet and the unique identifier of the terminal and the target learning duration set by the applet in advance.

Further, learning information of the students is collected by combining the target learning duration, and the method further comprises the following steps: the method comprises the steps that collected learning information and a pre-acquired operation identifier are sent to a server, and the server stores the learning information into a storage space corresponding to a terminal according to the operation identifier;

the method for acquiring the job identifier comprises the following steps:

and after the target learning duration is obtained, response information is sent to the server, the response information comprises the unique identifier of the terminal, the server generates an operation identifier corresponding to the unique identifier of the terminal according to the response information, and the operation identifier is sent to the terminal.

Further, the method further comprises:

and receiving and displaying a learning analysis result sent by the server, wherein the learning analysis result is obtained by analyzing the learning information by the server.

Further, the learning information includes sitting posture images of the student collected at least once within the target learning period;

the learning analysis result comprises sitting posture information of the student determined by the server according to the sitting posture image;

correspondingly, the learning analysis result sent by the server is received and displayed, and the method further comprises the following steps:

if the sitting posture information of the student does not meet the preset requirement, sending out prompt information;

and counting the total times of the prompt messages sent by the students during learning according to the target learning duration.

In a second aspect, a method for monitoring learning behavior is provided, which is applied to an applet, and includes:

acquiring a unique identifier of a terminal, and establishing a binding relationship according to the unique identifier of the terminal and account information of the applet;

and setting a target learning time length, sending the target learning time length to a terminal, and collecting learning information of students by the terminal according to the target learning time length.

Further, establishing a binding relationship according to the unique identifier of the terminal and the account information of the applet, including:

and sending a binding request to a server, wherein the binding request comprises the unique identifier of the terminal and the account information of the applet, and the server establishes a binding relationship between the account information of the applet and the unique identifier of the terminal according to the binding request.

Further, the sending the target duration to the terminal includes:

and after receiving a duration acquisition request sent by the terminal, the server returns the target learning duration corresponding to the unique identifier of the terminal to the terminal according to the duration setting instruction.

In a third aspect, a method for monitoring learning behaviors is provided, and is applied to a server, and the method includes:

receiving a binding request sent by the applet, wherein the binding request comprises a unique identifier of the terminal and account information of the applet, and establishing a binding relationship between the account information of the applet and the unique identifier of the terminal according to the binding request;

receiving a duration setting instruction sent by the applet, wherein the duration setting instruction comprises target learning duration and account information of the applet, and setting a corresponding relation between the account information of the applet and the target learning duration in the operation duration setting instruction;

receiving a duration obtaining request sent by a terminal, wherein the duration obtaining request comprises a unique identifier of the terminal, returning a target learning duration corresponding to the unique identifier of the terminal to the terminal according to the operation duration obtaining request, and collecting learning information of students by the terminal according to the target learning duration.

Returning the target learning duration corresponding to the unique identifier of the terminal to the terminal, and then:

receiving learning information collected by a terminal, and analyzing the learning information to obtain a learning analysis result;

and returning the learning analysis result to the terminal.

Further, returning the target learning duration corresponding to the unique identifier of the terminal to the terminal, and then:

receiving response information sent by a terminal, wherein the response information comprises a unique identifier of the terminal;

generating an operation identifier corresponding to the unique identifier of the terminal according to the response information, sending the operation identifier to the terminal, and returning the acquired learning information and the operation identifier by the terminal;

analyzing the learning information to obtain a learning analysis result, and then: and storing the learning analysis result to a storage space corresponding to the terminal according to the operation identification.

Further, the learning information comprises at least one sitting posture image of the student, which is acquired by the terminal within the target learning time length;

analyzing the learning information to obtain a learning analysis result, wherein the learning analysis result comprises: and determining the sitting posture information of the student according to the sitting posture image.

Further, the method for determining the sitting posture information of the student according to the sitting posture image comprises the following steps:

extracting the outlines of students from the sitting posture images to be used as outline maps;

and inputting the contour map into a pre-trained sitting posture determining model to obtain the sitting posture information of the student output by the sitting posture determining model.

Further, extracting the contour of the student from the sitting posture image as a contour map, comprising:

calculating the gray value of a pixel point in the sitting posture image to obtain a gray image;

converting the gray level image into a binary image;

performing edge tracking on the binary image to obtain edges in the binary image;

and taking the edge with the maximum number of pixel points as the outline of the student, and obtaining an outline image according to the pixel points on the outline of the student in the binary image.

Further, the sitting posture determining model comprises a sitting posture characteristic extraction sub-model and a sitting posture judging model;

inputting the contour map into a pre-trained sitting posture determination model to obtain sitting posture information of students output by the sitting posture determination model, wherein the sitting posture information comprises the following steps:

inputting the contour map into a sitting posture feature extraction model to obtain sitting posture features of students output by the sitting posture feature extraction model;

and inputting the sitting posture characteristics into the sitting posture judging model to obtain the sitting posture information of the students output by the sitting posture judging model.

Further, the sitting posture feature extraction model comprises a feature extraction layer and a classification layer;

the training method of the sitting posture feature extraction model comprises the following steps:

initializing parameters of a feature extraction layer and a classification layer;

the method comprises the following steps of taking contour diagrams of a certain number of sample students as training samples, taking sitting posture information of the sample students as sample labels, inputting the training samples and the sample labels to a feature extraction layer, and obtaining sitting posture features of the training samples output by the feature extraction layer;

inputting the sitting posture characteristics of the training samples into a classification layer to obtain the sitting posture prediction results of the training samples output by the classification layer;

and calculating the deviation between the sitting posture prediction result and the sample label corresponding to the training sample, and performing reverse feedback to adjust the parameters of the feature extraction layer and the classification layer until the deviation reaches a convergence condition, thereby obtaining a trained sitting posture feature extraction model.

Further, inputting the contour map into a pre-trained sitting posture feature extraction model to obtain sitting posture features of students output by the sitting posture feature extraction model, comprising:

and inputting the contour image into a sitting posture feature extraction model to obtain sitting posture features of the students output by a feature extraction layer of the sitting posture feature extraction model.

Further, the sitting posture judging model is a random forest classifier;

inputting the sitting posture characteristics into a pre-trained sitting posture judgment model, and obtaining the sitting posture information of students output by the sitting posture judgment model, wherein the sitting posture information comprises:

inputting the sitting posture characteristics into each decision tree of the random forest classifier to obtain the sitting posture information output by each decision tree, voting the sitting posture information output by all the decision trees by each decision tree, and determining the sitting posture information of the students according to the voting result.

Further, the training method of the random forest classifier comprises the following steps:

acquiring a training sample set, wherein training samples in the training sample set are sample contour diagrams carrying sitting posture information labels;

extracting n training samples from the training sample set in a random sampling mode to obtain n training sample subsets;

and correspondingly training a decision tree by utilizing each training sample subset to obtain the random forest separator with n decision trees.

In a fourth aspect, a terminal is provided, including:

the unique identifier sending module is used for displaying the unique identifier of the terminal, and the applet establishes a binding relationship with the account information of the applet according to the unique identifier of the terminal;

and the target time length acquisition module is used for acquiring the target learning time length set by the small program with the binding relationship and collecting the learning information of the student by combining the target learning time length.

Further, the target duration obtaining module comprises a duration obtaining submodule for obtaining the target learning duration set by the applet, and the duration obtaining submodule is used for receiving the target learning duration pushed by the applet with the binding relationship.

Further, the target duration obtaining module comprises a duration obtaining submodule for obtaining the target learning duration set by the applet, the duration obtaining submodule is used for sending a duration obtaining request to the server, the duration obtaining request comprises the unique identifier of the terminal, and the server returns the target learning duration corresponding to the unique identifier of the terminal to the terminal according to the duration obtaining request;

Further, the terminal further includes:

the learning information sending module is used for sending the collected learning information and the pre-acquired operation identification to the server, and the server stores the learning information into a storage space corresponding to the terminal according to the operation identification;

and the operation identification acquisition module is used for sending response information to the server after the target learning duration is acquired, wherein the response information comprises the unique identification of the terminal, and the server generates an operation identification corresponding to the unique identification of the terminal according to the response information and sends the operation identification to the terminal.

Further, the terminal further includes:

and the analysis result receiving module is used for receiving and displaying the learning analysis result sent by the server.

correspondingly, the terminal also comprises:

the prompting module is used for sending out prompting information if the sitting posture information of the student does not meet the preset requirement;

and the counting module is used for counting the total times of the prompt messages sent by the students according to the target learning duration.

In a fifth aspect, there is provided an applet comprising:

the unique identifier receiving module is used for acquiring the unique identifier of the terminal and establishing a binding relationship according to the unique identifier of the terminal and the account information of the applet;

and the target learning duration sending module is used for setting the target learning duration and sending the target learning duration to the terminal, and the terminal acquires the learning information of the students according to the target learning duration.

Further, the unique identifier receiving module comprises a binding establishment submodule for establishing a binding relationship according to the unique identifier of the terminal and the account information of the applet;

the binding establishment submodule is specifically configured to: and sending a binding request to a server, wherein the binding request comprises the unique identifier of the terminal and the account information of the applet, and the server establishes a binding relationship between the account information of the applet and the unique identifier of the terminal according to the binding request.

The target duration sending module is specifically configured to: and after receiving a duration acquisition request sent by the terminal, the server returns the target learning duration corresponding to the unique identifier of the terminal to the terminal according to the duration setting instruction.

In a sixth aspect, a server is provided, comprising:

the binding request receiving module is used for receiving a binding request sent by the applet, wherein the binding request comprises the unique identifier of the terminal and the account information of the applet, and the binding relationship between the account information of the applet and the unique identifier of the terminal is established according to the binding request;

the time length instruction receiving module is used for receiving a time length setting instruction sent by the small program, wherein the time length setting instruction comprises target learning time length and account information of the small program, and setting a corresponding relation between the account information of the small program and the target learning time length in the operation time length setting instruction;

the time length request receiving module is used for receiving a time length obtaining request sent by the terminal, the time length obtaining request comprises a unique identifier of the terminal, a target learning time length corresponding to the unique identifier of the terminal is returned to the terminal according to the operation time length obtaining request, and the terminal collects learning information of students according to the target learning time length.

Further, the server further comprises:

and the learning analysis module is used for receiving the learning information acquired by the terminal, analyzing the learning information to obtain a learning analysis result and returning the learning analysis result to the terminal.

Further, the server further comprises:

the operation identification generation module is used for receiving response information sent by the terminal, the response information comprises a unique identification of the terminal, an operation identification corresponding to the unique identification of the terminal is generated according to the response information, the operation identification is sent to the terminal, and the terminal returns the acquired learning information and the operation identification;

the learning analysis module is also used for storing the learning analysis result to a storage space corresponding to the terminal according to the operation identification.

the learning analysis module analyzes the learning information to obtain a learning analysis result, and the learning analysis result comprises the following steps: determining sitting posture information of the student according to the sitting posture image;

further, the learning analysis module includes:

the contour map extraction submodule is used for extracting the contour of the student from the sitting posture image to be used as a contour map;

and the sitting posture determining submodule is used for inputting the contour map into a pre-trained sitting posture determining model and obtaining the sitting posture information of the student output by the sitting posture determining model.

Further, the contour map extraction sub-module includes:

the gray image acquisition unit is used for calculating the gray value of a pixel point in the sitting posture image to obtain a gray image;

a binary image conversion unit for converting the grayscale image into a binary image;

the edge tracking unit is used for carrying out edge tracking on the binary image to obtain edges in the binary image;

and the contour map generating unit is used for taking the edge with the maximum number of pixel points as the contour of the student and obtaining the contour map according to the pixel points on the contour of the student in the binary image.

the sitting posture determining submodule comprises:

the sitting posture characteristic acquisition module is used for inputting the contour map into the sitting posture characteristic extraction model and acquiring the sitting posture characteristics of the students output by the sitting posture characteristic extraction model;

and the sitting posture information acquisition module is used for inputting the sitting posture characteristics into the sitting posture judgment model and acquiring the sitting posture information of the students output by the sitting posture judgment model.

the learning analysis module further comprises a first model training module for training the sitting posture feature extraction model, and the first model training module comprises: the parameter initialization submodule is used for initializing parameters of the feature extraction layer and the classification layer;

the training sample input submodule is used for taking the contour diagrams of a certain number of sample students as training samples, taking the sitting posture information of the sample students as sample labels, and inputting the training samples and the sample labels to the feature extraction layer to obtain the sitting posture features of the training samples output by the feature extraction layer;

the prediction result output submodule is used for inputting the sitting posture characteristics of the training samples into the classification layer to obtain the sitting posture prediction results of the training samples output by the classification layer;

and the parameter adjusting submodule is used for calculating the deviation between the sitting posture prediction result and the sample label corresponding to the training sample, and reversely feeding back and adjusting the parameters of the feature extraction layer and the classification layer until the deviation reaches a convergence condition, so as to obtain a trained sitting posture feature extraction model.

The sitting posture characteristic acquisition module is specifically used for: and inputting the contour image into a sitting posture feature extraction model to obtain sitting posture features of the students output by a feature extraction layer of the sitting posture feature extraction model.

Further, the sitting posture judging model is a random forest classifier;

the sitting posture information acquisition module is specifically used for: inputting the sitting posture characteristics into each decision tree of the random forest classifier to obtain the sitting posture information output by each decision tree, voting the sitting posture information output by all the decision trees by each decision tree, and determining the sitting posture information of the students according to the voting result.

Further, the learning analysis module further comprises a second model training module for training the sitting posture judgment model, and the second model training module comprises:

the sample acquisition submodule is used for acquiring a training sample set, and training samples in the training sample set are sample contour diagrams carrying sitting posture information labels;

the sample subset acquisition module is used for extracting n training samples from the training sample set in a random sampling mode to obtain n training sample subsets;

and the decision tree training module is used for correspondingly training a decision tree by utilizing each training sample subset to obtain the random forest separator with n decision trees.

In a seventh aspect, an electronic device is provided, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, the steps of the method as provided in the first, second, or third aspect are implemented.

In an eighth aspect, a computer-readable storage medium is provided, having stored thereon a computer program which, when executed by a processor, carries out the steps of the method as provided in the first, second or third aspect.

According to the monitoring method for the learning behaviors, the terminal, the small program and the server, the small program can establish a binding relationship according to the unique identifier of the terminal and the account information of the small program by displaying the unique identifier of the terminal, then parents set the target learning duration by using the convenience of the small program for use immediately, the terminal collects the learning information of students according to the target learning duration set by the parents and supervises the students to complete learning tasks within the set time, and the embodiment of the application utilizes the convenience of the small program to be in linkage interaction with the terminal, helps the parents to supervise and manage the learning conditions of the students, is beneficial to improving the work completion efficiency of the students, develops a good time management concept, and helps the parents to know the learning conditions of children.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings used in the description of the embodiments of the present application will be briefly described below.

Fig. 1 is a schematic flowchart of a learning supervision system according to an embodiment of the present application;

fig. 2 is a schematic flowchart of a method for monitoring learning behaviors according to an embodiment of the present disclosure;

FIG. 3 is a schematic diagram of learning information displayed by a terminal after a student completes learning according to an embodiment of the application;

fig. 4 is an interaction diagram illustrating sitting posture monitoring between a terminal and a server according to an embodiment of the present application;

fig. 5 is an interaction diagram illustrating a batch operation between a terminal and a server according to an embodiment of the present application;

fig. 6 is a schematic flowchart of a method for monitoring learning behaviors according to another embodiment of the present application;

fig. 7 is a flowchart illustrating a method for monitoring learning behaviors according to still another embodiment of the present application;

fig. 8 is a schematic flow chart illustrating a process of obtaining sitting posture information from a sitting posture image by a server according to an embodiment of the present application;

fig. 9 is a schematic structural diagram of a terminal according to an embodiment of the present application;

FIG. 10 is a schematic diagram of an applet according to an embodiment of the present application;

fig. 11 is a schematic structural diagram of a server according to an embodiment of the present application;

fig. 12 is an interaction diagram of a terminal, an applet, and a server according to an embodiment of the present application;

fig. 13 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

Reference will now be made in detail to the embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are exemplary only for the purpose of explaining the present application and are not to be construed as limiting the present invention.

As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or wirelessly coupled. As used herein, the term "and/or" includes all or any element and all combinations of one or more of the associated listed items.

To make the objects, technical solutions and advantages of the present application more clear, embodiments of the present application will be described in further detail below with reference to the accompanying drawings.

Referring to fig. 1, a monitoring method in which a terminal, a server, and an applet collectively perform a learning behavior is taken as an example. The learning supervision system provided by the embodiment of the application comprises an applet 10, a server 11, a terminal 12 and the like; the applet refers to an application program which can be used without downloading and installing and is realized based on a host program, the host program can be QQ, WeChat and the like, as is well known, common application programs all need to download an installation package, for example, an android system, files in an apk format need to be downloaded, and the installation of the application program can be realized by running the files in the apk format, but the applet does not need to be downloaded and installed at all, for example, the applet in the WeChat is a browser and server structure. The structure is more convenient compared with a C/S structure (the C/S structure refers to a Server-Client machine, namely a Client-Server (C/S) structure), because an installation package does not need to be downloaded to upgrade the Client, the wechat plays a role of a browser, the applet is placed in the Server of the remote wechat, when the user needs to use the applet, the user directly sends a request in the wechat, and the remote Server provides service, therefore, when the applet is used, only a key point needs to be clicked, the applet is interacted with information of the remote host, and the applet can be used after the interaction is successful. The server 11 may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing cloud computing services. The terminal 12 may include a mobile phone, a learning machine, an early education machine, a monitor, a smart television, a tablet Computer, a notebook Computer, or a Personal Computer (PC), etc. The terminal 10 may further have a client, which may be an application client or a browser client, etc., for students to learn. The applet 10 and the server 11 are connected through a network, such as a wired or wireless network, etc., the server 11 and the terminal 12 are connected through a network, such as a wired or wireless network, etc., and the applet 10 and the terminal are connected through a network, such as a wired or wireless network, etc.

In an embodiment, the terminal 12 may display its own unique identifier in the two-dimensional code, so that the applet 10 obtains the unique identifier of the terminal by scanning the two-dimensional code, when the terminal is a Mobile phone terminal of an android system, an International Mobile Equipment Identity (IMEI) may be used as the unique identifier, for an android tablet computer without a call function, a serial number of a product may be used as the unique identifier, for an apple Mobile phone, a serial number of the Mobile phone may be used as the unique identifier, and the like. The applet 10 stores account information which can be information of parents of students or information of students, the applet 10 sends the account information of the applet and the unique identifier of the terminal 12 to the server 11, the server 11 binds the unique identifier of the terminal 12 with the account information of the applet 10, the server returns the binding relationship to the applet 10, after the applet 10 confirms that the binding relationship is established, the method comprises the steps that a target learning duration is sent to a server 11, the server 11 establishes a corresponding relation between the target learning duration and a unique identifier of an applet 10, a duration obtaining request is sent to the server 11 by a terminal 12, the duration obtaining request comprises the unique identifier of the terminal 12, the server 11 returns the target learning duration corresponding to the unique identifier of the terminal 12 to the terminal 12 according to the duration obtaining request, and the terminal 12 collects learning information of students by combining the target learning duration. The learning information of the embodiment of the application may include homework completed by a student, the homework may be obtained by taking a picture when the homework is a paper homework, the homework may be obtained according to an operation result of the student on the exercise on the terminal when the homework is an electronic homework, for example, an actual learning duration may be determined by a time when the student submits the homework, when the homework is a paper homework, a time uploaded by taking a picture is taken as an end time of the actual learning duration, when the homework is an electronic homework, an operation (submission) time of the student on a last question is taken as an end time of the actual learning duration, the learning information may further include sitting posture information of the student, a number of times and a duration of leaving a monitoring range, and the like, generally, the terminal may turn on a camera function of the terminal when the student is learning, and supervise the learning behavior of the student, if the student leaves the shooting collection range, the student is determined to leave the monitoring range.

The method for monitoring the learning behavior can be applied to various scenes needing to supervise the learning of students, for example, when homework needs to be done, parents, students or teachers can set the target learning duration through the applet by the method provided by the embodiment, the terminals bound with the applet monitor the homework of the students according to the target learning duration, and the learning information of the students is collected, so that the parents can conveniently supervise the homework doing duration of the students.

The embodiment of the present application can be applied to monitoring the learning behaviors of students by parents at home, and is also applicable to monitoring the learning behaviors of students by teachers in classrooms, it should be understood that the usage scenarios of classrooms described in the embodiment of the present application are different from the usage scenarios of the classrooms in that the monitored objects are more, for example, in an optional embodiment, a monitor serving as a terminal is installed in a teacher, the learning information collected by the monitor can be the sitting posture, the sitting posture holding duration, whether to meet the ears during the learning process, and the like of the students, and the teachers set the target learning duration by a small program, for example, a class hour: and after the study class begins, the monitor collects study information in combination with the target study duration, each student can be further identified in a face identification mode, and then the study information corresponding to each student is sent to the applet, so that a teacher can conveniently master the study information of each student.

The execution method of the server in the embodiment of the application can be completed in a cloud computing (cloud computing) mode, and the cloud computing is a computing mode, and distributes computing tasks on a resource pool formed by a large number of computers, so that various application systems can obtain computing power, storage space and information service according to needs. The network that provides the resources is referred to as the "cloud". Resources in the "cloud" appear to the user as being infinitely expandable and available at any time, available on demand, expandable at any time, and paid for on-demand.

As a basic capability provider of cloud computing, a cloud computing resource pool (called as an ifas (Infrastructure as a Service) platform for short is established, and multiple types of virtual resources are deployed in the resource pool and are selectively used by external clients.

According to the logic function division, a PaaS (Platform as a Service) layer can be deployed on an IaaS (Infrastructure as a Service) layer, a SaaS (Software as a Service) layer is deployed on the PaaS layer, and the SaaS can be directly deployed on the IaaS. PaaS is a platform on which software runs, such as a database, a web container, etc. SaaS is a variety of business software, such as web portal, sms, and mass texting. Generally speaking, SaaS and PaaS are upper layers relative to IaaS.

The embodiment of the application provides a method for monitoring a learning behavior, which is applied to a terminal, and as shown in fig. 2, the method includes:

s101, displaying the unique identifier of the terminal, and establishing a binding relationship between the applet and the account information of the applet according to the unique identifier of the terminal.

Specifically, the unique identifier of the terminal itself may be recorded in the two-dimensional code in advance, and then the two-dimensional code is displayed on a display screen or a body of the terminal, so that the applet obtains the unique identifier of the terminal itself in a manner of scanning the two-dimensional code, for example, the unique identifier of the terminal itself is recorded in a preset URL (Uniform Resource Locator), and the applet obtains the unique identifier of the terminal by accessing the URL.

The terminal of the embodiment of the application lays a foundation for binding the small program and the terminal by showing the unique identifier of the terminal.

S102, acquiring target learning duration set by the small program with the binding relationship, and collecting learning information of students according to the target learning duration.

It should be understood that, in one embodiment, the parent sets a target learning period, i.e., a planned period of a one-time learning process, which is also a time set for improving the time management ability of the student, through the applet. The terminal receives the target learning duration set by the applet with the binding relationship, so that the learning information of the student is collected according to the target learning duration. The problem may be sent to the terminal by the applet, or may be downloaded from the database by the terminal, and the application is not limited further. The learning information refers to information generated by the student in the learning process, such as homework, completion time, sitting posture, half-time rest time, number of times, and the like, which are completed by the student in the learning process.

According to the monitoring method for the learning behaviors, the binding relationship can be established by the applet according to the unique identifier of the terminal and the account information of the applet through the unique identifier of the display terminal, then the convenience of the applet for use is utilized by the parents to set the target learning duration, the terminal collects the learning information of the student according to the target learning duration set by the parents, and the student is urged to complete the learning task within the specified time.

On the basis of the foregoing embodiments, there are two ways for the terminal of the embodiment of the present application to obtain the target learning duration set by the applet, and in actual use, the target learning duration may be obtained in any one of the following two ways:

(1) receiving a target learning duration pushed by an applet having a binding relationship.

Specifically, when the device running the applet and the terminal are in the same local area network or are in communication connection through bluetooth, the applet can directly push the target learning duration to the terminal without using a server for message transfer. The method is suitable for parents to be in a close distance with students or to be used when the network is disconnected (the network is commonly known that the terminal and/or equipment running the small program cannot be in communication connection with the Internet), and the parents can set the target learning duration in real time through the small program to be referred by the terminal.

(2) Sending a time length obtaining request to a server, wherein the time length obtaining request comprises a unique identifier of a terminal, and the server returns a target learning time length corresponding to the unique identifier of the terminal to the terminal according to the time length obtaining request;

In the method, the terminal and the applet interact with each other through the server serving as the middleware, on the basis, specifically, the applet firstly sends account information of the applet and the unique identifier of the terminal to the server, the server establishes a binding relationship between the applet and the terminal, and the server can send notification information of successful binding to the terminal and the applet after establishing the binding relationship, so that the applet sends the set target learning duration to the server after learning that the binding is successful. The server stores the target learning duration set by the applet, and then the terminal sends a duration acquisition request to the server, wherein the duration acquisition request is used for indicating that the terminal wants to acquire the target learning duration.

The unique identifier of the terminal is set in the time length obtaining request, so that the server firstly determines the applet determined by the unique identifier according to the unique identifier of the terminal recorded in the time length obtaining request, then further determines the target learning time length set by the applet, and the server returns the searched target learning time length to the terminal, so that the terminal obtains the target learning time length corresponding to the unique identifier of the terminal.

It should be understood that the duration obtaining request includes a request identifier in addition to the unique identifier of the terminal, each request has a corresponding unique identifier, that is, a request identifier, and the server stores the request identifier of each request in advance, so that the purpose of the request can be determined after the request is received.

On the basis of the above embodiments, as an optional embodiment, the terminal collects learning information of the student according to the target learning duration, and then further includes: and sending the acquired learning information and the pre-acquired operation identification to a server, and storing the learning information to a storage space corresponding to the terminal by the server according to the operation identification.

After the learning information of the students is collected, the terminal of the embodiment of the application can send the learning information and the homework identification to the server, and the server stores the learning information in the storage space corresponding to the terminal, so that the ordered storage of the learning information is realized.

Before the learning information is sent to the server, the method also involves the step that the terminal acquires the job identification from the server, specifically:

the terminal sends response information to the server after acquiring the target learning duration, the response information comprises the unique identifier of the terminal, the server generates an operation identifier corresponding to the unique identifier of the terminal according to the response information, and the operation identifier is sent to the terminal.

The purpose of the job identifier of the embodiment of the present application is to establish association between the learning information and the terminal, and therefore the job identifier may be variable or fixed. When the job identifier is changed, different job identifiers indicate different learning conditions or different jobs, for example, the learning conditions (jobs) of today and tomorrow belong to two different learning stages, so the server can set one job identifier for the learning condition of today and set another job identifier for the learning condition of tomorrow.

On the basis of the foregoing embodiments, as an optional embodiment, the method further includes the following steps of collecting learning information of a student according to a target learning duration:

the collected learning information is sent to a server, the server analyzes the learning information to obtain a learning analysis result, and the learning analysis result is returned to the terminal;

and receiving and displaying the learning analysis result.

As can be seen from the above description, the learning information in the embodiment of the present application refers to information generated by a student during a learning process, for example, a homework, a completion time, a sitting posture, a time of half break, a number of times of half-time break, and the like of the student during the learning process, and the server analyzes the learning information, for example, may include calculating a correct rate of the homework, a completion speed, whether the sitting posture is standard, and the like. The terminal sends the acquired learning information to the server, and the server generates a learning analysis result, so that the operation overhead of the terminal can be reduced, and the hardware requirement on the terminal is reduced.

The learning information collected by the embodiment of the application comprises the actual learning duration, further, the server or the terminal determines whether the learning duration of the student meets the requirement or not by comparing the size relation between the actual learning duration and the target learning duration, the analyzed conclusion is sent to the terminal for displaying as a learning analysis result, and the server or the terminal can also determine the reward information for the student according to the learning analysis result.

For example, if the learning task is to do a job, the parent sets the target learning period in order to expect the student to be able to finish more efficiently, and therefore the actual learning period is preferably shorter than the target learning period, and therefore the actual learning period is most rewarded if it does not exceed the target learning period, and the actual learning period is less rewarded if it exceeds the target learning period, the more the time is exceeded.

Table 1 is a table of correspondence between learning completion time and rewards in the embodiment of the present application, and as shown in table 1, the rewards in the embodiment of the present application are saffron, which is a virtual article, and the longer the actual learning duration is, the lower the learning efficiency of the student is, the fewer the corresponding rewards are, and the saffron can be used for the student to exchange prizes later, such as famous teacher explanation, classical test questions, and the like.

Learning completion time	The number of small safflower
		Is done regularly	4
Is finished within 20 percent of time	2
		Overtime is more than 20%	0

TABLE 1 learning finish time and reward corresponding relation table

If the learning task is a preview or review, the goal of the parent setting the target learning duration is to expect that the student will spend more time to complete, so the actual learning duration is preferably longer than the target learning duration, so the actual learning duration is awarded the least if it just reaches the target learning duration, and the actual learning duration is awarded the more if it exceeds the target learning duration.

In addition, the reward of the embodiment of the present application can also be determined by combining the work accuracy and the learning duration, for example, the work accuracy is firstly divided into two grades: excellence and goodness, learning duration is also divided into two cases: the operation is finished on time and the learning time is finished on time, the highest-grade small red flowers are rewarded, for example, 4 small red flowers are rewarded if the operation accuracy is excellent and the learning time is finished on time, the second highest-grade small red flowers are rewarded, for example, 2 small red flowers are rewarded if the operation accuracy is excellent and the learning time is finished on time, and the small red flowers are not rewarded if the operation accuracy is excellent and the learning time is finished on time.

The learning information of the terminal collection of this application still includes the position of sitting image of student when studying, through gathering position of sitting information, can form good position of sitting custom for helping the student. After receiving the learning information containing the sitting posture image, the terminal determines the sitting posture information of the student according to the sitting posture image, and the sitting posture information can be a conclusion whether the sitting posture is standard or not.

Correspondingly, the receiving and displaying of the learning analysis result by the terminal further comprises:

the server of the embodiment of the application can classify the sitting posture information in advance, the correct sitting posture meeting the preset requirements is determined, if the student in the sitting posture image collected by the terminal does not belong to the correct sitting posture, the sitting posture information of the student is nonstandard or not meeting the preset requirements, the terminal can send the reminding information after receiving the learning analysis result, the reminding information can be pop-up window display prompt information on the terminal, alarm sound can also be sent, the reminding information can also be sent to a small program, the server can understand that when sending the reminding information to the small program, parents can also receive irregular reminding of the sitting posture of children through the small program. This is not particularly limited by the examples of the present application.

Furthermore, after the student finishes learning, the total times of the reminding information sent out in the learning process is counted, rewards corresponding to different times can be pre-established, and the size of the rewards is inversely proportional to the times. Table 2 is a corresponding relationship table of the reminding times and the rewards in the embodiment of the present application, as shown in table 2, the rewards in the embodiment of the present application are saffron, which is a virtual article, and the less the sitting posture reminding is, the better the sitting posture of the student is kept, the more standard the corresponding rewards are, and the more saffron can be used for the student to exchange prizes, such as famous teacher explanation, classical test questions and the like.

Sitting posture reminder (second time)	Small red flower (number)
		0	5
1	4
		2	3
3	2
		4	1
5	0

TABLE 2 reminding times and reward corresponding relation table

Fig. 3 is a schematic view of learning information displayed by a terminal after the student finishes learning in the embodiment of the present application, and as shown in fig. 3, the embodiment of the present application can count the time of actual learning and the number of times of sitting posture reminding of the student, and can also display inspirational words, thereby helping the student improve self-confidence and increasing the learning love degree.

Fig. 4 is an interaction schematic diagram of monitoring sitting postures of a terminal and a server according to the embodiment of the application, as shown in fig. 4, the terminal periodically collects sitting posture images of students in a target learning duration and uploads the collected sitting posture images to the server in real time, the server analyzes the sitting posture images to obtain sitting posture information of the students and returns the sitting posture information to the terminal, after the terminal receives the sitting posture information, if the sitting posture information is found not to meet preset requirements, the terminal reminds the students to correct the sitting postures, after the students finish learning, the terminal counts the total times of sitting posture reminding and reports the counted total times to the server, and the server determines corresponding rewards according to the total times.

Fig. 5 is an interaction schematic diagram illustrating that a terminal and a server perform job correction in an embodiment of the present application, and as shown in fig. 5, the terminal collects jobs completed by students by photographing and uploads the collected jobs to the server, and the server automatically corrects the jobs by using a job correction interface provided by a current cloud manufacturer (for example, Tencent cloud), obtains scores of the students, and returns the scores and rewards to the terminal for display according to a preset corresponding relationship between the scores and the rewards (for example, safflowers). Taking the currently realized intelligent education solution of Tencent cloud as an example, the operation correction interface such as oral evaluation, composition correction, mathematic operation correction and the like is provided at present, wherein the composition correction interface supports two modules of overall comment and sentence-by-sentence comment, the detail dimension relates to terms such as vocabulary, parts of speech, sentence patterns, chapter structures, content relevance and the like, modification suggestions are given, and the capability of correcting English by images or texts can be rapidly developed and realized by accessing the interface. The mathematic homework correcting interface supports various oral arithmetic subject types including vertical type, detachable type, four rules and the like, and can be widely applied to the primary school math teaching scene. The interface is accessed, so that the ability of correcting the primary school quick calculation work can be quickly developed, for example, the paper marking interface based on the picture text recognition function of the test paper can be used for the scenes of marking the paper for the primary school, the junior high school and the high school, such as the mathematics, the Chinese and the English.

The present application also provides a method for monitoring learning behavior, which is applied to an applet, as shown in fig. 6, and includes:

s201, acquiring the unique identifier of the terminal, and establishing a binding relationship according to the unique identifier of the terminal and the account information of the applet.

Specifically, the terminal may record the unique identifier of the terminal in the two-dimensional code in advance, and then display the two-dimensional code on a display screen or a body of the terminal, so that the applet obtains the unique identifier of the terminal by scanning the two-dimensional code, for example, the unique identifier of the terminal is recorded in a preset URL (Uniform Resource Locator), and the applet obtains the unique identifier of the terminal by accessing the URL. The applet in the embodiment of the application can establish the binding relationship with the terminal by acquiring the unique identifier of the terminal.

S202, setting a target learning duration, sending the target learning duration to a terminal, and collecting learning information of students by the terminal according to the target learning duration.

It should be understood that the parent sets a target learning period, i.e., a planned period of one-time learning process, which is also a time set for improving the time management ability of the student, through the applet. The terminal receives the target learning duration set by the applet with the binding relationship, so that the learning information of the student is collected according to the target learning duration. The exercises can be sent to the terminal by the applet or downloaded from the exercise database by the terminal, and the application is not further limited. The learning information refers to information generated by the student in the learning process, such as a correct rate, a completion speed, a sitting posture, a time and a number of times of rest in the middle of the learning process of the student, and the specific content of the learning information is not further limited in the embodiment of the present application.

According to the monitoring method for the learning behaviors, the unique identification of the terminal is obtained through the small program, the small program can establish a binding relationship according to the unique identification of the terminal and the account information of the small program, then the parents can set the target learning duration by using the convenience of the small program which is ready to use, the learning information of students is collected by the terminal according to the target learning duration set by the parents, and the students are urged to finish learning tasks within the specified time.

On the basis of the foregoing embodiments, the establishing, by the applet, a binding relationship with account information of the applet according to the unique identifier of the terminal includes:

It should be noted that the applet records the account information of the applet and the unique identifier of the terminal in the binding request and sends the binding request to the server, the server establishes the binding relationship between the applet and the terminal, and the server can send a notification message of successful binding to the terminal and the applet after establishing the binding relationship, so that the applet obtains the result of successful binding.

On the basis of the foregoing embodiments, as an optional embodiment, the sending the target duration to the terminal includes:

The method comprises the steps that after learning that binding is successful, the applet sends set target learning time length to a server, the target learning time length is recorded in a time length setting instruction, and account information of the applet is also included in the time length setting instruction, so that the server establishes a corresponding relation of the terminal-the applet-the target learning time length after receiving the time length setting instruction, and returns the corresponding target learning time length to the terminal after receiving a time length obtaining instruction sent by the terminal.

The present application further provides a method for monitoring learning behaviors, which is applied to a server, and as shown in fig. 7, the method includes:

s301, receiving a binding request sent by the applet, wherein the binding request comprises a unique identifier of the terminal and account information of the applet, and establishing a binding relationship between the account information of the applet and the unique identifier of the terminal according to the binding request;

s302, receiving a duration setting instruction sent by the applet, wherein the duration setting instruction comprises a target learning duration and account information of the applet, and setting a corresponding relation between the account information of the applet and the target learning duration in the operation duration setting instruction;

s303, receiving a duration obtaining request sent by the terminal, wherein the duration obtaining request comprises a unique identifier of the terminal, returning a target learning duration corresponding to the unique identifier of the terminal to the terminal according to the operation duration obtaining request, and collecting learning information of students by the terminal according to the target learning duration.

The server of the embodiment of the application establishes the binding relationship between the small program and the terminal by receiving the binding request of the small program, then receives the time setting instruction sent by the small program, establishes the corresponding relationship between the small program and the target learning time, and can give play to the corresponding target learning time to the terminal according to the corresponding relationship between the small program, the terminal and the target learning time after receiving the time obtaining request sent by the terminal, so that the student learning time is remotely set by a household through the small program, and the burden of the household is relieved.

On the basis of the foregoing embodiments, as an optional embodiment, returning a target learning duration corresponding to the unique identifier of the terminal to the terminal, and then further including:

and returning the learning analysis result to the terminal.

The learning information in the embodiment of the present application refers to information generated by a student in a learning process, for example, homework, completion time, sitting posture, time and number of half-time breaks of the student in the learning process, and the specific content of the learning information is not further limited in the embodiment of the present application. The server analyzes the learning, which may include, for example, calculating a correct rate of the job, a completion rate, whether the sitting posture is standard, and the like. The terminal sends the acquired learning information to the server, and the server generates a learning analysis result, so that the operation overhead of the terminal can be reduced, and the hardware requirement on the terminal is reduced.

On the basis of the foregoing embodiments, as an optional embodiment, the server returns, to the terminal, the target learning duration corresponding to the unique identifier of the terminal, and then further includes:

and generating a job identifier corresponding to the unique identifier of the terminal according to the response information, sending the job identifier to the terminal, and returning the acquired learning information and the job identifier by the terminal.

Correspondingly, the server analyzes the learning information to obtain a learning analysis result, and the learning analysis result comprises the following steps: and storing the learning analysis result to a storage space corresponding to the terminal according to the operation identification.

After the learning information of the students is collected, the terminal of the embodiment of the application can send the learning information and the homework identification to the server, and the server stores the learning information in the storage space corresponding to the terminal, so that the ordered storage of the learning information is realized. Optionally, when the server establishes the correspondence between the applet and the terminal, a storage space dedicated for storing the terminal may be set in the database according to the unique identifier of the terminal.

On the basis of the above embodiments, as an alternative embodiment, the learning information includes at least one sitting posture image of the student collected by the terminal within the target learning time period.

The terminal of the embodiment of the application can randomly collect the sitting posture images of the students in the target learning time length and can also periodically collect the sitting posture images of the students according to the preset time period, and the application is not specifically limited. The sitting posture image is an image for recording the sitting posture of the student. Specifically, after the terminal collects images of students during learning, noise in the images can be removed by using an image filtering technology, then a target area can be located, the target area can be a head area, specifically, hair or five sense organs can be understood to locate the head area, and then a human body area is framed by taking the target area as a center to obtain a sitting posture image.

Correspondingly, the server analyzes the learning information to obtain a learning analysis result, and the learning analysis result comprises the following steps: and determining the sitting posture information of the student according to the sitting posture image.

The learning information of the terminal collection of this application still includes the position of sitting image of student when study, and the server can help the student to form good position of sitting custom through carrying out the analysis to the position of sitting image and obtaining student's position of sitting information. On the basis of the above embodiment, as an alternative embodiment, the determining the sitting posture information of the student according to the sitting posture image includes:

s401, extracting the outline of the student from the sitting posture image to be used as an outline map.

The embodiment of the application also relates to extracting the outline of the student from the sitting posture image, obtaining the outline map, and determining the sitting posture of the student by using the outline of the student.

Specifically, step S401 includes:

s4011, calculating gray values of pixel points in the sitting posture image to obtain a gray image.

This application can set up the weight that the colour value of red channel, green channel and blue channel corresponds according to the relative size of people's eye to red, green and blue sensitivity, then to the RGB color space of position of sitting image, carries out the weighted summation to every pixel point at the colour value of red channel, green channel and blue channel, and as the grey level of pixel point, specifically can calculate the grey level according to following formula:

g(x,y)＝W_RR+W_GG+W_BB

wherein, W_R、W_GAnd W_BRespectively representing the weights corresponding to the color values of the red channel, the green channel and the blue channel; r, G and B respectively identify the color values of the pixel points in the red channel, the green channel and the blue channel; x and y respectively represent x-axis and y-axis coordinates of the pixel points, and g (x, y) represents the gray value of the pixel points. Optionally, W_R＝0.30，W_G＝0.59，W_B＝0.11。

And S4012, converting the gray level image into a binary image.

After obtaining the gray image, a maximum inter-class difference method can be adopted to obtain a binary image, and the binary image is divided into a background part and an object part according to the gray characteristic of the image. The larger the inter-class variance between the background and the object, the larger the difference between the two parts constituting the image, and the smaller the difference between the two parts when part of the object is mistaken for the background or part of the background is mistaken for the object. Thus, a segmentation that maximizes the inter-class variance means that the probability of false positives is minimized. The maximum inter-class difference method can obtain a threshold value T which enables the background image and the foreground image to be best segmented, for a pixel point with a gray value larger than the threshold value T, the gray value of the pixel point in the binary image is 255, and the pixel point is white, and if the gray value is not larger than the threshold value T, the gray value is 0, and the pixel point is black.

S4013, performing edge tracking on the binary image to obtain edges in the binary image.

Specifically, the embodiment of the application can detect all edges of a binary image by using OpenCV, where OpenCV is a cross-platform computer vision and machine learning software library which includes a large number of related functions for image processing, and the application can check the edge of a miniature by using an equivalent function built in OpenCV, and it is understood that the edge is a junction between an image region and another attribute region and is a place where a region attribute changes suddenly.

S4014, the edge with the maximum number of the pixel points is used as the contour of the student, and a contour map is obtained according to the pixel points on the contour of the student in the binary image.

It should be understood that after the edge corresponding to the student contour is determined, the binary image may be marked according to the coordinates of each pixel point of the edge, then the area surrounded by the edge is used as the student contour, and the area not belonging to the student contour in the binary image is deleted, that is, the contour map is obtained.

S402, inputting the contour map into a pre-trained sitting posture determining model to obtain sitting posture information of the student output by the sitting posture determining model.

According to the sitting posture information obtaining method and device, sitting posture information is obtained through a machine learning mode, specifically, a contour diagram is input into a pre-trained sitting posture determining and extracting model, sitting posture characteristics of students output by the sitting posture characteristic extracting model are obtained, and the sitting posture information of the students is determined through excellent computing capacity and understanding capacity of machine learning.

It should be understood that before step S402 is executed, a sitting posture determination model may be trained in advance, and specifically, the sitting posture determination model may be obtained by the following method: first, a certain number of sample sitting posture images are collected, a contour map of each sample sitting posture image is obtained, and sitting posture information of each sample sitting posture image is determined, wherein the sitting posture information can be two-classification results, such as a correct sitting posture and an incorrect sitting posture, or multi-classification results, such as a standard sitting posture, a stooped sitting posture, a shoulder inclined sitting posture and the like. And randomly training an initial model based on the outline graph of the sample sitting posture image and the sitting posture information of the sample sitting posture image so as to obtain a sitting posture determination model, wherein the initial model can be a single neural network model or a combination of a plurality of neural network models. The sitting posture determining model can obtain sitting posture information according to the input contour map.

The model structure of the sitting posture determining model in the embodiment of the present application may adopt a Convolutional Neural Network (CNN), a Generative Adaptive Network (GAN), and the like, and the embodiment of the present application does not specifically limit the model structure of the sitting posture determining model.

According to the sitting posture image acquisition method and device, the sitting posture image of the student is acquired within the target learning time length, the outline of the student is further extracted from the sitting posture image and is used as the outline map, the outline map is input into the pre-trained sitting posture determination model, the sitting posture information of the student is obtained, parents can be replaced to supervise the sitting posture of the student personally, and the sitting posture of the student is determined in a neural network mode, so that the accuracy is higher than that of manual detection. In addition, the contour map can be directly sent to the small program, so that a captain can directly see the screenshot of the sitting posture of the student through the small program.

On the basis of the above embodiments, as an optional embodiment, the sitting posture determining model includes a sitting posture feature extraction sub-model and a sitting posture judgment model;

s501, inputting the contour map into the sitting posture characteristic extraction model to obtain the sitting posture characteristics of the students output by the sitting posture characteristic extraction model.

Specifically, the sitting posture feature extraction model comprises a feature extraction layer and a classification layer;

s5011, initializing parameters of a feature extraction layer and a classification layer;

s5012, using contour maps of a certain number of sample students as training samples, using sitting posture information of the sample students as sample labels, and inputting the training samples and the sample labels into the feature extraction layer to obtain sitting posture features of the training samples output by the feature extraction layer;

s5013, inputting the sitting posture characteristics of the training samples into a classification layer to obtain a sitting posture prediction result of the training samples output by the classification layer;

s5014, calculating deviation between the sitting posture prediction result and a sample label corresponding to the training sample, and reversely feeding back and adjusting parameters of the feature extraction layer and the classification layer until the deviation reaches a convergence condition to obtain a trained sitting posture feature extraction model.

The model structure of the sitting posture feature extraction model in the embodiment of the application can be a deep convolutional neural network, and the deep convolutional neural network mainly comprises a convolutional layer, a pooling layer, a full-link layer and a logistic regression layer, and specifically comprises the following steps:

and (3) rolling layers: the convolution operation essentially represents the input in another way, and the training of the entire network is the intermediate parameters needed to train out this representation. Convolution is an analytical calculation in mathematics, and generally performs a convolution operation on an image matrix and a 2-dimensional matrix to obtain a new matrix, for example, an original image matrix is:

by convolution kernels

The original image matrix is convolved, and then,

the convolution operation results obtained are:

a pooling layer: the method is a special processing operation for data in a convolutional neural network, reduces the characteristic size of a picture through pooling processing, and can effectively solve the problem of large calculation amount caused by taking the result of the previous layer as input.

Full connection layer: the layer is the layer with the most consumed parameters in the network, if the input of the fully connected layer is 4 x 100 and the output of the fully connected layer is 512, the layer needs 4 x 100 x 512 parameters; in a typical convolutional layer, if the convolution kernel is 4 × 4 and the output is 512, then only 4 × 512 parameters are needed. The output of the full connection layer corresponds to the output of the classification number.

And a logistic regression layer, which is a Learning algorithm for solving a Supervised Learning (Supervised Learning) problem, for performing logistic regression with the purpose of minimizing an error between a tag value of training data and a predicted value, and capable of outputting a probability result corresponding to each component.

Therefore, the convolution layer, the pooling layer and the full-connection layer in the deep convolutional neural network can be used as the feature extraction layer, and the logistic regression layer is used as the classification layer.

On the basis of the above embodiments, as an alternative embodiment, inputting the contour map into a sitting posture feature extraction model trained in advance, and obtaining the sitting posture feature of the student output by the sitting posture feature extraction model, the method includes:

And S502, inputting the sitting posture characteristics into the sitting posture judgment model to obtain the sitting posture information of the student output by the sitting posture judgment model.

The sitting posture judging model of the embodiment of the application can adopt a random forest classifier, and the random forest classifier refers to a classifier which trains and predicts a sample by utilizing a plurality of trees. Through a plurality of tests, the random forest classifier is found to classify the sitting posture characteristics, and compared with the classification result of a classification layer carried by a convolutional neural network, the classification result is more accurate.

The training method of the random forest classifier comprises the following steps:

s5021, a training sample set is obtained, and training samples in the training sample set are sample contour diagrams carrying sitting posture information labels;

s5022, extracting n training samples from the training sample set in a random sampling mode to obtain n training sample subsets;

s5023, correspondingly training a decision tree by utilizing each training sample subset to obtain the random forest separator with n decision trees.

The sitting posture determining model further comprises two sub-models, wherein the first sub-model is used for outputting sitting posture characteristics according to the contour map, the second sub-model is used for obtaining sitting posture information according to the sitting posture characteristics, a convolutional neural network is used as a sitting posture characteristic extracting model for obtaining the sitting posture characteristics, a random forest classifier is used as a sitting posture judging model, and the accuracy of the sitting posture determining model is higher than that of a single neural network model when the same sample set is used.

Fig. 8 is a schematic flow chart of the server according to the embodiment of the present application for obtaining sitting posture information according to the sitting posture image, as shown in fig. 8,

s601, receiving a sitting posture image sent by a terminal;

s602, preprocessing the sitting posture image, wherein the preprocessing process comprises the following steps: filtering the image, positioning a head area in the filtered image, framing a human body area by taking the head area as a center, and obtaining the outline of a student as an outline map in an edge tracking mode;

s603, extracting the features of the contour map, specifically realizing the features through a convolutional neural network, inputting the contour map into a convolutional layer of the convolutional neural network for convolution processing, inputting a convolution result into a pooling layer, reducing the feature size, and inputting a result output by the pooling layer into a full-link layer to obtain the sitting posture features output by the full-link layer.

And S604, inputting the sitting posture characteristics into the random forest classifier to obtain sitting posture information output by the random forest classifier.

An embodiment of the present application provides a terminal, as shown in fig. 9, the terminal may include: a unique identifier sending module 101 and a target duration obtaining module 102, wherein,

the unique identifier sending module 101 is used for displaying the unique identifier of the terminal, and the applet establishes a binding relationship with the account information of the applet according to the unique identifier of the terminal;

the target duration obtaining module 102 is configured to obtain a target learning duration set by the applet having the binding relationship, and collect learning information of the student according to the target learning duration.

The terminal provided in the embodiment of the present invention specifically executes the process of the method embodiment on the terminal side, and please refer to the content of the embodiment of the learning behavior monitoring method on the terminal side for details, which is not described herein again. According to the terminal provided by the embodiment of the invention, the binding relationship can be established by the applet according to the unique identifier of the terminal and the account information of the applet by displaying the unique identifier of the terminal, so that parents can set the target learning duration by using the convenience of the instant start and use of the applet, the terminal collects the learning information of students according to the target learning duration set by the parents, and the students are supervised and urged to complete the learning task within the specified time.

On the basis of the foregoing embodiments, as an optional embodiment, the target duration obtaining module includes a duration obtaining submodule configured to obtain a target learning duration set by the applet, and the duration obtaining submodule is configured to receive the target learning duration pushed by the applet having a binding relationship.

On the basis of the above embodiments, as an optional embodiment, the target duration obtaining module includes a duration obtaining submodule for obtaining a target learning duration set by the applet, the duration obtaining submodule is configured to send a duration obtaining request to the server, the duration obtaining request includes a unique identifier of the terminal, and the server returns the target learning duration corresponding to the unique identifier of the terminal to the terminal according to the duration obtaining request;

On the basis of the foregoing embodiments, as an optional embodiment, the terminal further includes:

On the basis of the above embodiments, as an alternative embodiment, the learning information includes at least one sitting posture image of the student collected within the target learning duration;

correspondingly, the terminal also comprises:

and the counting module is used for counting the total times of the prompt messages sent by the students during learning according to the target learning duration.

An embodiment of the present application provides an applet, as shown in fig. 10, the applet may include: a unique identification receiving module 201 and a target duration transmitting module 202, wherein,

the unique identifier receiving module 201 is used for acquiring a unique identifier of the terminal and establishing a binding relationship according to the unique identifier of the terminal and account information of the applet;

and the target learning duration sending module 202 is configured to set a target learning duration, send the target learning duration to the terminal, and the terminal collects learning information of the student according to the target learning duration.

The applet provided in the embodiment of the present invention specifically executes the flow of the method embodiment on the applet side, and please refer to the content of the embodiment of the learning behavior monitoring method on the applet side for details, which is not described herein again. According to the applet provided by the embodiment of the invention, the binding relationship can be established by the applet according to the unique identifier of the terminal and the account information of the applet by acquiring the unique identifier of the terminal, so that parents can set the target learning duration by using the convenience of the applet, the terminal collects the learning information of students according to the target learning duration set by the parents, and the students are urged to complete the learning task within the specified time.

On the basis of the above embodiments, as an optional embodiment, the unique identifier receiving module includes a binding establishment sub-module for establishing a binding relationship between the unique identifier of the terminal and the account information of the applet, and the binding establishment sub-module is specifically configured to: and sending a binding request to a server, wherein the binding request comprises the unique identifier of the terminal and the account information of the applet, and the server establishes a binding relationship between the account information of the applet and the unique identifier of the terminal according to the binding request.

An embodiment of the present application provides a server, as shown in fig. 11, the server may include: a binding request receiving module 301, a duration instruction receiving module 302, and a duration request receiving module 303, wherein,

a binding request receiving module 301, configured to receive a binding request sent by an applet, where the binding request includes a unique identifier of a terminal and account information of the applet, and establish a binding relationship between the account information of the applet and the unique identifier of the terminal according to the binding request;

a duration instruction receiving module 302, configured to receive a duration setting instruction sent by the applet, where the duration setting instruction includes a target learning duration and account information of the applet, and sets a corresponding relationship between the account information of the applet and the target learning duration in the operation duration setting instruction;

the duration request receiving module 303 is configured to receive a duration obtaining request sent by the terminal, where the duration obtaining request includes a unique identifier of the terminal, return a target learning duration corresponding to the unique identifier of the terminal to the terminal according to the operation duration obtaining request, and collect learning information of a student according to the target learning duration.

The server provided in the embodiment of the present invention specifically executes the process of the method embodiment on the server side, and please refer to the content of the embodiment of the learning behavior monitoring method on the server side for details, which is not described herein again. The server provided by the embodiment of the invention establishes the binding relationship between the applet and the terminal by receiving the binding request of the applet, then receives the time length setting instruction sent by the applet, establishes the corresponding relationship between the applet and the target learning time length, and can give play to the corresponding target learning time length to the terminal according to the corresponding relationship between the applet, the terminal and the target learning time length after receiving the time length obtaining request sent by the terminal, so that the student learning time length can be remotely set by the household manager in a applet mode, and the burden of the household manager is reduced.

On the basis of the foregoing embodiments, as an optional embodiment, the server further includes:

On the basis of the above embodiments, as an optional embodiment, the learning information includes at least one sitting posture image of the student, which is acquired by the terminal within the target learning duration;

on the basis of the above embodiments, as an alternative embodiment, the learning analysis module includes:

On the basis of the above embodiments, as an alternative embodiment, the contour map extraction sub-module includes:

the sitting posture determining submodule comprises:

On the basis of the above embodiments, as an optional embodiment, the sitting posture feature extraction model includes a feature extraction layer and a classification layer;

On the basis of the above embodiments, as an optional embodiment, the sitting posture judging model is a random forest classifier;

On the basis of the foregoing embodiments, as an optional embodiment, the learning analysis module further includes a second model training module for training the sitting posture judgment model, and the second model training module includes:

In the embodiment of the application, the server may be logically divided into 4 services, which are a gateway service, a duration management service, an account service and a database service, respectively, where the gateway service is configured to perform data transmission with the applet and the terminal, the duration management service is configured to manage information related to learning duration, the account service is configured to associate information related to an account/a unique identifier of the applet/the terminal, and the database is configured to store information related to learning content, specifically, an interaction process of the server, the applet and the terminal may be shown in fig. 12, as shown in fig. 12, the terminal (referred to as a device side in fig. 12) displays the unique identifier, that is, a device DSN data (Source name), the applet sends the device DSN of the device side and account information of the applet to the gateway service, the gateway server sends the account information of the device DSN and the applet to the account service, the account service writes the account information of the equipment DSN and the small program into the database, after the database is successfully written, the information of successful binding is returned to the account service, and the information is further returned to the small program through the gateway service. When the working time length is set, the applet sends a time length setting instruction for setting the working time length to the gateway service, the grid server sends the time length setting instruction to the time length management service, the time length management server sets the working tomato clock time length according to the time length setting instruction, it should be understood that the tomato clock time length is the time length set according to a tomato working method. When the device side supervises learning, firstly, a duration acquisition request for checking the operation duration is sent to the gateway service, the duration management server applies a check binding relationship to the account service according to the duration acquisition request, determines the tomato clock duration set by the applet bound with the device side, returns the tomato clock duration to the device side, the device side starts the supervision operation completion time according to the tomato clock duration, then the account service checks the binding relationship again, and the bound account information is returned to the duration management service, the duration management service generates the unique identifier of the operation, the duration management service returns the unique identifier of the operation to the equipment end, meanwhile, the unique identifier is inserted into a job basic information table of the database, and the job basic information table is used for recording basic information of the job (such as an account number of the applet, the unique identifier of the equipment terminal, the job duration and the like); when the equipment submits the homework, the finished time length and the sitting posture image are sent to the time length management service, the time length management service reviews the homework and calculates the reward, the reward in the application is the number of safflowers, the homework score and the reward are inserted into a required homework completion record table according to the unique identification of the homework, the homework completion record table is used for recording detailed learning information, besides the information recorded in the homework basic information table, the correction result and the reward of the learning are further recorded, and the time length management service returns a reply language with encouraging property to the equipment end, so that the interest of the student in learning is promoted.

An embodiment of the present application provides an electronic device, including: a memory and a processor; at least one program stored in the memory for execution by the processor, which when executed by the processor, implements: the binding relation can be established according to the unique identification of the terminal and the account information of the small program by the small program, then a parent sets the target learning duration by using the convenience of the small program which is immediately used, the terminal collects the learning information of students according to the target learning duration set by the parent, and the students are supervised and urged to complete the learning task within the specified time.

In an alternative embodiment, there is provided an electronic device, as shown in fig. 13, the electronic device 4000 shown in fig. 13 comprising: a processor 4001 and a memory 4003. Processor 4001 is coupled to memory 4003, such as via bus 4002. Optionally, the electronic device 4000 may further comprise a transceiver 4004. In addition, the transceiver 4004 is not limited to one in practical applications, and the structure of the electronic device 4000 is not limited to the embodiment of the present application.

The Processor 4001 may be a CPU (Central Processing Unit), a general-purpose Processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other Programmable logic device, a transistor logic device, a hardware component, or any combination thereof. Which may implement or perform the various illustrative logical blocks, modules, and circuits described in connection with the disclosure. The processor 4001 may also be a combination that performs a computational function, including, for example, a combination of one or more microprocessors, a combination of a DSP and a microprocessor, or the like.

Bus 4002 may include a path that carries information between the aforementioned components. The bus 4002 may be a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, or the like. The bus 4002 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in FIG. 13, but this is not intended to represent only one bus or type of bus.

The Memory 4003 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical Disc storage, optical Disc storage (including Compact Disc, laser Disc, optical Disc, digital versatile Disc, blu-ray Disc, etc.), a magnetic Disc storage medium or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to these.

The memory 4003 is used for storing application codes for executing the scheme of the present application, and the execution is controlled by the processor 4001. Processor 4001 is configured to execute application code stored in memory 4003 to implement what is shown in the foregoing method embodiments.

The present application provides a computer-readable storage medium, on which a computer program is stored, which, when running on a computer, enables the computer to execute the corresponding content in the foregoing method embodiments. Embodiments of the present application also provide a computer program product or computer program comprising computer instructions stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions to cause the computer device to perform the learning behavior monitoring method provided in the various alternative implementations of the above aspects. Compared with the prior art, the binding relation can be established according to the unique identification of the terminal and the account information of the small program by the small program through the unique identification of the display terminal, then the parents set the target learning duration by utilizing the convenience of the small program which is immediately used, the learning information of students is collected by the terminal according to the target learning duration set by the parents, and the students are supervised and urged to complete the learning task within the specified time.

It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and may be performed in other orders unless explicitly stated herein. Moreover, at least a portion of the steps in the flow chart of the figure may include multiple sub-steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed alternately or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

The foregoing is only a partial embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.

Claims

1. A method for monitoring learning behaviors is applied to a terminal, and is characterized in that the method comprises the following steps:

displaying the unique identification of the terminal, and establishing a binding relationship according to the unique identification of the terminal and account information of the applet;

and acquiring the target learning duration set by the small program with the binding relationship, and collecting the learning information of the students by combining the target learning duration.

2. The learning supervision method according to claim 1, wherein the learning information of the student is collected in combination with the target learning duration, and thereafter the method further comprises: the acquired learning information and the acquired operation identification are sent to a server, and the server stores the learning information to a storage space corresponding to the terminal according to the operation identification;

the method for acquiring the job identifier comprises the following steps:

and after the target learning duration is obtained, sending response information to the server, wherein the response information comprises the unique identifier of the terminal, and the server generates an operation identifier corresponding to the unique identifier of the terminal according to the response information and sends the operation identifier to the terminal.

3. The method for monitoring learning behavior of claim 2, further comprising:

4. A method for monitoring learning activities as claimed in claim 3, wherein the learning information comprises at least one sitting posture image of the student collected within the target learning period;

the receiving and displaying server analyzes the learning information to obtain a learning analysis result, and the method further comprises the following steps:

and counting the total times of the prompt messages sent by the students according to the target learning duration.

5. A method for monitoring learning behavior, applied to an applet, the method comprising:

and setting a target learning duration, sending the target learning duration to the terminal, and collecting learning information of students by combining the target learning duration through the terminal.

6. A method for monitoring learning behaviors, which is applied to a server, is characterized by comprising the following steps:

receiving a binding request sent by an applet, wherein the binding request comprises a unique identifier of a terminal and account information of the applet, and establishing a binding relationship between the account information of the applet and the unique identifier of the terminal according to the binding request;

receiving a duration setting instruction sent by the applet, wherein the duration setting instruction comprises a target learning duration and account information of the applet, and setting a corresponding relation between the account information of the applet and the target learning duration in the operation duration setting instruction;

receiving a duration obtaining request sent by the terminal, wherein the duration obtaining request comprises the unique identifier of the terminal, returning a target learning duration corresponding to the unique identifier of the terminal to the terminal according to the operation duration obtaining request, and collecting learning information of students by the terminal in combination with the target learning duration.

7. The method for monitoring learning behaviors of claim 6, wherein the step of returning the target learning duration corresponding to the unique identifier of the terminal to the terminal further comprises the step of:

the method comprises the steps of receiving learning information collected by a terminal, analyzing the learning information to obtain a learning analysis result, and returning the learning analysis result to the terminal.

8. The method for monitoring learning behaviors of claim 7, wherein the step of returning the target learning duration corresponding to the unique identifier of the terminal to the terminal further comprises the step of:

receiving response information sent by the terminal, wherein the response information comprises a unique identifier of the terminal, generating an operation identifier corresponding to the unique identifier of the terminal according to the response information, sending the operation identifier to the terminal, and returning the acquired learning information and the operation identifier by the terminal;

the analyzing the learning information to obtain a learning analysis result, and then further comprising: and storing the learning analysis result to a storage space corresponding to the terminal according to the operation identification.

9. The method for monitoring learning behaviors of claim 7, wherein the learning information comprises at least one sitting posture image of the student collected by the terminal in the target learning duration;

the analyzing the learning information to obtain a learning analysis result includes: and determining the sitting posture information of the student according to the sitting posture image.

10. A terminal, comprising:

11. An applet, comprising:

the unique identifier receiving module is used for acquiring a unique identifier of a terminal and establishing a binding relationship according to the unique identifier of the terminal and the account information of the applet;

and the target learning duration sending module is used for setting a target learning duration and sending the target learning duration to the terminal, and the terminal collects learning information of students by combining the target learning duration.

12. A server, comprising:

the binding request receiving module is used for receiving a binding request sent by an applet, wherein the binding request comprises a unique identifier of a terminal and account information of the applet, and the binding relationship between the account information of the applet and the unique identifier of the terminal is established according to the binding request;

a duration instruction receiving module, configured to receive a duration setting instruction sent by the applet, where the duration setting instruction includes a target learning duration and account information of the applet, and sets a corresponding relationship between the account information of the applet and the target learning duration in the operation duration setting instruction;

the time length request receiving module is used for receiving a time length obtaining request sent by the terminal, the time length obtaining request comprises the unique identifier of the terminal, the target learning time length corresponding to the unique identifier of the terminal is returned to the terminal according to the operation time length obtaining request, and the terminal collects learning information of students by combining the target learning time length.

13. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method for monitoring learning behavior according to any of claims 1 to 9 are implemented when the program is executed by the processor.

14. A computer-readable storage medium, characterized in that it stores computer instructions which cause the computer to perform the steps of the method for monitoring learning behavior according to any one of claims 1 to 9.

15. A computer program, characterized in that the computer program comprises computer instructions stored in a computer-readable storage medium, which, when read by a processor of a computer device from the computer-readable storage medium, cause the processor to execute the computer instructions, causing the computer device to perform the steps of the method for monitoring learning behavior according to any one of claims 1 to 9.