Child learning posture adjusting method and system based on active learning

Abstract

The invention discloses a child learning posture adjusting method and system based on active learning, and relates to the technical field of posture recognition, and the child learning posture adjusting method based on active learning provided by the invention is characterized in that through multi-dimensional data collection, a scientific posture evaluation model and an intelligent correction mechanism, the posture recognition accuracy is improved; real-time monitoring, rationality evaluation and dynamic adjustment of learning postures of children are realized; according to the method, bad learning postures can be accurately recognized, personalized correction can be carried out according to individual historical data, misjudgment is avoided, children are helped to keep healthy learning postures, and body injuries caused by long-term bad postures are prevented; besides, the system guides the children to adjust the posture in time through real-time feedback and alarm prompt, and meanwhile, the system has accuracy, adaptability and expandability; according to the whole scheme, the learning posture adjusting efficiency and the intelligent level can be improved, and scientific guarantee is provided for healthy cultivation of long-term learning habits of children.

Description

Technical Field

[0001] The present invention relates to the technical field of posture recognition, and in particular to a method and system for adjusting children's learning posture based on active learning. Background Art

[0002] At present, children's health problems caused by long-term poor learning postures are becoming increasingly prominent, such as myopia, spinal curvature and other posture-related diseases. Although there are some posture monitoring and correction devices on the market, most of them have a single function and cannot take into account the multi-dimensional collection, real-time evaluation and intelligent adjustment of posture at the same time. In addition, these traditional devices often lack personalized design and cannot provide accurate guidance for different children's body shapes and posture habits.

[0003] With the development of sensor technology, computer vision and artificial intelligence technology, posture analysis methods based on multi-sensor fusion and deep learning are gradually emerging. These technologies can collect and analyze multi-dimensional posture data, combine historical data for judgment, and provide new technical means for posture monitoring and adjustment. However, there is currently a lack of a systematic and intelligent solution that can efficiently complete the whole process management of children's learning posture, including data collection, posture evaluation, state classification and real-time correction. Based on this, the present invention provides a method and system for adjusting children's learning posture based on active learning. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a method and system for adjusting children's learning posture based on active learning, which solves the problems in the background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for adjusting children's learning posture based on active learning, comprising:

[0006] Step 1: When the child is in a learning state, collect the child's posture data through a variety of sensors and cameras;

[0007] Step 2: Evaluate the children’s learning posture based on their posture data and determine the overall rationality score;

[0008] Step 3: According to the overall rationality score, the children's learning posture status is classified into standard learning posture status, good learning posture status, learning posture status that needs to be improved, and unreasonable learning posture status;

[0009] Step 4: When the learning posture state is a state where the learning posture needs to be improved or a state where the learning posture is unreasonable, a dynamic correction analysis is performed based on the previously acquired historical posture data of the child, and whether the current learning posture is the child's posture habit is determined according to the analysis result; and according to the determination result, whether to trigger an alarm message to remind the child;

[0010] Step 5: When the alarm message is triggered to prompt the child according to step 4, a warning message is issued to prompt the child to adjust the learning posture.

[0011] As a further solution of the present invention: in the step 1, the child posture data includes: sitting pressure data, distance data and joint angle data;

[0012] The joint angle data include a pitch angle, a yaw angle, a roll angle, an angle between the neck and the upper back, an angle between the upper back and the lower back, a shoulder angle and an elbow bending angle.

[0013] As a further solution of the present invention: in step 2, the specific method of evaluating the child's learning posture according to the child's posture data is:

[0014] AS1: Label the child's posture data, respectively: Sitting posture pressure data is labeled , distance data is marked as , the pitch angle, yaw angle, roll angle, neck-upper back angle, upper back-lower back angle, shoulder angle and elbow flexion angle are marked as ,in, ;

[0015] AS2: Normalize the 9 data in the child posture data of AS1 and map all data to The normalization formula for interval is as follows:

[0016] in, Represented as the corresponding data in the child posture data, , The sequence corresponds to the sitting pressure data, distance data, pitch angle, yaw angle, roll angle, angle between neck and upper back, angle between upper back and lower back, shoulder angle and elbow bending angle in the child posture data. is the reasonable interval of the corresponding data, It is represented as the normalized result of the corresponding data in the children's posture data;

[0017] Finally, the normalized data ;

[0018] AS3: Set weights for the corresponding data in the child posture data to indicate their influence on the rationality of the posture. The weights are set as follows:

[0019] Sitting pressure data weight , distance data weight , the weight of joint angle data is , and the sum of the joint angle data weights is ;

[0020] AS4: Establish a posture scoring model, and then use the following scoring function to reflect the posture scoring model;

[0021] Reasonableness score of sitting pressure data : Determine ideal sitting posture pressure data based on children's body shape , determine the score as:

[0022] Distance data plausibility score :The ideal distance data is determined by the child's body shape , determine the score as:

[0023] Joint angle data rationality score :

[0024] Set each joint angle data The reasonable range is , determine the ideal angle based on the child's body shape , determine the score as:

[0025] Rationality score of comprehensive joint angle data :

[0026] Then the overall rationality score is obtained by the following formula: :

[0027] In the formula, , and are weight coefficients respectively.

[0028] As a further solution of the present invention: in the step 3, the specific method of classifying the learning posture state of the child according to the overall rationality score to obtain the standard learning posture state, the good learning posture state, the learning posture state that needs to be improved and the unreasonable learning posture state is:

[0029] If the overall plausibility score , indicating the standard state of learning posture;

[0030] If the overall plausibility score , indicating a good learning posture;

[0031] If the overall plausibility score , indicating that the learning posture needs to be improved;

[0032] If the overall plausibility score , indicating that the learning posture is unreasonable.

[0033] As a further solution of the present invention: in the step 4, the dynamic correction analysis is performed based on the pre-acquired historical posture data of the child, and whether the current posture behavior is the posture habit of the child is determined according to the analysis result; according to the judgment result, the specific content of determining whether to trigger the alarm information to prompt the child is:

[0034] BS1: Obtain children’s historical posture data and record children’s historical posture data within a specific time window:

[0035] in, Indicates time The historical posture data of children at all times, including: historical sitting posture pressure data , historical distance data ;Historical joint angle data ; is the length of a specific time window;

[0036] BS2: Statistical analysis of children’s historical posture data to build a mean model for children :

[0037] BS3: Next, calculate the variance model of the child’s historical posture data :

[0038] BS4: The child's posture data after obtaining the overall rationality score at the current moment is:

[0039] in, The child's posture data collected at the current moment, including: current sitting posture pressure data , Current distance data ; Current joint angle data ;

[0040] Calculate the current child posture data and the mean model of the child The deviation of :

[0041] Then the normalized deviation vector is determined by the following formula :

[0042] Quantifying overall deviation using Euclidean distance :

[0043] In the formula, represents the normalized bias vector The Euclidean norm of is calculated by summing the squares of each component of the deviation vector and taking the square root; Each component in Indicates the deviation size of a single feature; represents the normalized bias vector The kth component of ; The 9 in the figure refers to the 9 children’s posture data collected, including: sitting pressure data; distance data; pitch angle, yaw angle, roll angle, angle between neck and upper back, angle between upper back and lower back, shoulder angle and elbow bending angle;

[0044] Set dynamic thresholds based on the statistical characteristics of children's historical posture data :

[0045] In the formula, is an empirical parameter used to balance the detection of normal fluctuations and abnormal postures; Expressed as The overall deviation of the moment.

[0046] As a further solution of the present invention: the specific content of determining whether to trigger the alarm information to remind the child also includes:

[0047] To make a correction judgment:

[0048] when : Indicates that the current learning posture is the child’s posture habit, and the current posture is judged to be reasonable;

[0049] when : Indicates that the current learning posture is not the child’s posture habit and triggers an alarm prompt directly.

[0050] A children's learning posture adjustment system based on active learning, comprising:

[0051] A data collection module is used to collect children's posture data through a variety of sensors and cameras when the children are in a learning state;

[0052] A posture assessment module, used for assessing the children's learning posture according to the children's posture data, and determining an overall rationality score;

[0053] The posture judgment module is used to classify the learning posture status of children according to the overall rationality score, and obtain the learning posture standard status, the learning posture good status, the learning posture needs to be improved status and the learning posture unreasonable status;

[0054] The intelligent correction module is used to perform dynamic correction analysis based on the previously acquired historical posture data of the child when the learning posture state is a state where the learning posture needs to be improved or a state where the learning posture is unreasonable, and to determine whether the current learning posture is the posture habit of the child according to the analysis result; and to determine whether to trigger an alarm message to remind the child according to the judgment result;

[0055] The warning module is used to send out a warning message to prompt the child to adjust the posture when the intelligent correction module determines that the alarm information is triggered to prompt the child.

[0056] The present invention provides a method and system for adjusting children's learning posture based on active learning. Compared with the prior art, it has the following beneficial effects:

[0057] The active learning-based children's learning posture adjustment method provided by the present invention realizes real-time monitoring, rationality evaluation and dynamic adjustment of children's learning posture through multi-dimensional data collection, scientific posture evaluation model and intelligent correction mechanism. The method can not only accurately identify bad learning postures, but also perform personalized corrections based on individual historical data to avoid misjudgment, help children maintain a healthy learning posture, and prevent physical injuries caused by long-term bad postures.

[0058] In addition, the system guides children to adjust their posture in time through real-time feedback and alarm prompts, while being accurate, adaptable and scalable. The overall solution can not only improve the efficiency and intelligence level of learning posture adjustment, but also provide scientific guarantee for the healthy cultivation of children's long-term learning habits. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The present invention will be further described below in conjunction with the accompanying drawings.

[0060] Figure 1 A flowchart of the steps of a method for adjusting children's learning posture based on active learning according to the present invention;

[0061] Figure 2 The present invention is a structural framework diagram of a children's learning posture adjustment system based on active learning. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] Embodiment 1

[0064] See also Figure 1 ,The present invention provides a method for adjusting children's learning posture based on active learning, comprising:

[0065] Step 1: When the child is in a learning state, collect the child's posture data through a variety of sensors and cameras;

[0066] The child posture data includes: sitting posture pressure data, distance data and joint angle data;

[0067] The joint angle data include pitch angle, yaw angle, roll angle, angle between neck and upper back, angle between upper back and lower back, shoulder angle and elbow bending angle;

[0068] Specifically, the sensor includes a pressure sensor and a distance sensor. The pressure sensor is used to collect sitting pressure data: the pressure sensor is installed on the seat to obtain the sitting pressure data; the distance sensor is used to collect the distance data between the head and the desktop: the distance sensor is installed at a fixed position on the desktop to obtain the distance data;

[0069] Use a camera to capture the child's sitting video stream, identify the child's skeleton points in real time through a deep learning model (such as ResNet or YOLO series networks), and determine the joint angle data based on the skeleton points, including head angles: pitch angle, yaw angle, roll angle; back angles: the angle between the neck and the upper back, and the angle between the upper back and the lower back; arm angles: shoulder angle, elbow bending angle;

[0070] It should be noted that: using a deep learning model to identify the child's bone points in real time and determining the joint angle data based on the bone points is an existing technology and will not be elaborated here;

[0071] Step 2: Evaluate the children’s learning posture based on their posture data and determine the overall rationality score;

[0072] The specific method of evaluating the child's learning posture according to the child's posture data is as follows:

[0073] AS1: Label the child's posture data, respectively: Sitting posture pressure data is labeled , distance data is marked as , the pitch angle, yaw angle, roll angle, neck-upper back angle, upper back-lower back angle, shoulder angle and elbow flexion angle are marked as ,in, ;

[0074] AS2: Normalize the 9 data in the child posture data of AS1 and map all data to The normalization formula for interval is as follows:

[0075] in, Represented as the corresponding data in the child posture data, , The sequence corresponds to the sitting pressure data, distance data, pitch angle, yaw angle, roll angle, angle between neck and upper back, angle between upper back and lower back, shoulder angle and elbow bending angle in the child posture data. is the reasonable interval of the corresponding data, It is represented as the normalized result of the corresponding data in the children's posture data;

[0076] Finally, the normalized data ;

[0077] It should be noted that the reasonable range is further explained as follows: The reasonable interval is the distance data The reasonable range can be set to 15cm-25cm. The specific reasonable range is set in advance by professional staff, which is the preset value.

[0078] AS3: Set weights for the corresponding data in the child posture data to indicate their influence on the rationality of the posture. The weights are set as follows:

[0079] Sitting pressure data weight , distance data weight , the weight of joint angle data is , and the sum of the joint angle data weights is ; The value of the weight is set by professional staff;

[0080] AS4: Establish a posture scoring model, and then use the following scoring function to reflect the posture scoring model;

[0081] Sitting pressure data rationality score : Determine ideal sitting posture pressure data based on children's body shape Specific parameters are set by professionals based on the child's body shape, and the scores are determined as follows:

[0082] Distance data plausibility score :The ideal distance data is determined by the child's body shape Specific parameters are set by professionals based on the child's body shape, and the scores are determined as follows:

[0083] Joint angle data rationality score :

[0084] Set each joint angle data The reasonable range is , determine the ideal angle based on the child's body shape Specific parameters are set by professionals based on the child's body shape, and the scores are determined as follows:

[0085] Comprehensive joint angle data rationality score :

[0086] Then the overall rationality score is obtained by the following formula: :

[0087] In the formula, , and are weight coefficients respectively;

[0088] The multi-dimensional posture data is quantified by normalization and weight distribution. Through a scientific scoring model and reasonable interval setting, the rationality of the learning posture is accurately evaluated. It can not only reflect the different effects of various characteristics on the posture, but also adapt to the individual differences of children, providing a reliable basis for classification and correction.

[0089] Step 3: According to the overall rationality score, the children's learning posture status is classified into standard learning posture status, good learning posture status, learning posture status that needs to be improved, and unreasonable learning posture status;

[0090] The specific method of classifying the learning posture status of children according to the overall rationality score to obtain the standard learning posture status, good learning posture status, learning posture status that needs to be improved, and unreasonable learning posture status is as follows:

[0091] If the overall plausibility score , indicating the standard state of learning posture;

[0092] If the overall plausibility score , indicating that the learning posture is in good condition, but needs further improvement;

[0093] If the overall plausibility score , indicating that the learning posture needs to be improved;

[0094] If the overall plausibility score , indicating that the learning posture is unreasonable;

[0095] Step 4: When the learning posture state is a state where the learning posture needs to be improved or a state where the learning posture is unreasonable, a dynamic correction analysis is performed based on the previously acquired historical posture data of the child, and whether the current learning posture is the child's posture habit is determined according to the analysis result; and according to the determination result, whether to trigger an alarm message to remind the child;

[0096] The dynamic correction analysis is performed based on the previously acquired historical posture data of the child, and the specific content of determining whether the current learning posture behavior is the posture habit of the child is determined according to the analysis result; and determining whether to trigger the alarm information to prompt the child according to the judgment result is:

[0097] BS1: Obtain children’s historical posture data and record children’s historical posture data within a specific time window:

[0098] in, Indicates time The historical posture data of children at all times, including: historical sitting posture pressure data , historical distance data ;Historical joint angle data ; Joint angle; is the length of a specific time window;

[0099] BS2: Statistical analysis of children’s historical posture data to build a mean model for children :

[0100] BS3: Next, calculate the variance model of the child’s historical posture data :

[0101] BS4: The child's posture data after obtaining the overall rationality score at the current moment is:

[0102] in, The child's posture data collected at the current moment, including: current sitting posture pressure data , Current distance data ; Current joint angle data ;

[0103] Calculate the child's posture data at the current moment and the child's mean mode The deviation of :

[0104] Next, determine the normalized deviation vector , normalized by the variance of historical data:

[0105] Quantifying overall deviation using Euclidean distance :

[0106] In the formula, represents the normalized bias vector The Euclidean norm (L2 norm) of is calculated by summing the squares of each component of the deviation vector and taking the square root; Each component in Indicates the deviation size of a single feature, which has the same unit and scale after being standardized; represents the normalized bias vector The kth component of ; The 9 in the figure refers to the 9 posture feature data collected, including: sitting pressure data; distance data; pitch angle, yaw angle, roll angle, angle between neck and upper back, angle between upper back and lower back, shoulder angle and elbow bending angle;

[0107] Set dynamic thresholds based on the statistical characteristics of children's historical posture data :

[0108] In the formula, It is an empirical parameter (usually between 1.5 and 2), which is set by professional staff to balance the detection of normal fluctuations and abnormal postures; Expressed as Overall deviation of the moment;

[0109] Then make a correction judgment:

[0110] when : Indicates that the current learning posture is the child’s posture habit, and the current posture is judged to be reasonable;

[0111] when : Indicates that the current learning posture is not the child’s posture habit, and an alarm prompt is triggered directly;

[0112] It should be noted that after obtaining the rationality score, if the score belongs to the state that the learning posture needs to be improved or the learning posture is unreasonable, the system needs to further analyze the source of the score deviation. At this time, by comparing the current child posture data with the historical data, it is determined whether the low score is caused by the system's misjudgment;

[0113] The mean and variance models are established through the historical posture data of children, and the Euclidean distance and dynamic threshold are combined to achieve personalized correction and anomaly detection, optimize the false alarm rate and dynamically adjust the algorithm to ensure continuous guidance of healthy posture and intelligent adaptation in complex scenarios.

[0114] Step 5: When the intelligent correction module determines that the alarm information is triggered to prompt the child, a warning message is issued to prompt the child to adjust the posture.

[0115] Embodiment 2

[0116] See also Figure 2 In the specific implementation process, this embodiment is based on the first embodiment and is different from the first embodiment in that this embodiment further provides a children's learning posture adjustment system based on active learning, including:

[0117] A data collection module is used to collect children's posture data through a variety of sensors and cameras when the children are in a learning state;

[0118] A posture assessment module, used for assessing the children's learning posture according to the children's posture data, and determining an overall rationality score;

[0119] The posture judgment module is used to classify the learning posture status of children according to the overall rationality score, and obtain the learning posture standard status, the learning posture good status, the learning posture needs to be improved status and the learning posture unreasonable status;

[0120] The intelligent correction module performs dynamic correction analysis based on the previously acquired historical posture data of the child when the learning posture state is a state where the learning posture needs to be improved or a state where the learning posture is unreasonable, and determines whether the current learning posture is the child's posture habit based on the analysis result; and determines whether to trigger an alarm message to remind the child based on the judgment result;

[0121] The warning module is used to send out a warning message to prompt the child to adjust the posture when the intelligent correction module determines that the alarm information is triggered to prompt the child.

[0122] Embodiment 3

[0123] The specific implementation process of this embodiment includes the entire implementation process of the above two groups of embodiments.

[0124] Some of the data in the above formulas are numerically calculated by taking their dimensions. At the same time, the contents not described in detail in this specification belong to the existing technologies well known to those skilled in the art.

[0125] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for adjusting children's learning posture based on active learning, characterized in that: include: Step 1: When the child is in a learning state, collect the child's posture data through a variety of sensors and cameras; Step 2: Evaluate the children’s learning posture based on their posture data and determine the overall rationality score; Step 3: According to the overall rationality score, the children's learning posture status is classified into standard learning posture status, good learning posture status, learning posture status that needs to be improved, and unreasonable learning posture status; Step 4: When the learning posture state is a state where the learning posture needs to be improved or a state where the learning posture is unreasonable, a dynamic correction analysis is performed based on the previously acquired historical posture data of the child, and whether the current learning posture is the child's posture habit is determined according to the analysis result; and according to the determination result, whether to trigger an alarm message to remind the child; Step 5: When the alarm message is triggered to prompt the child according to step 4, a warning message is issued to prompt the child to adjust the learning posture.

2. A method for adjusting children's learning posture based on active learning according to claim 1, characterized in that: In the step 1, the child posture data includes: sitting pressure data, distance data and joint angle data; The joint angle data include a pitch angle, a yaw angle, a roll angle, an angle between the neck and the upper back, an angle between the upper back and the lower back, a shoulder angle and an elbow bending angle.

3. A method for adjusting children's learning posture based on active learning according to claim 2, characterized in that: In step 2, the specific method of evaluating the child's learning posture according to the child's posture data is: AS1: Label the child's posture data, respectively: Sitting posture pressure data is labeled , distance data is marked as , the pitch angle, yaw angle, roll angle, neck-upper back angle, upper back-lower back angle, shoulder angle and elbow flexion angle are marked as ,in, ; AS2: Normalize the 9 data in the child posture data of AS1 and map all data to The normalization formula for interval is as follows: ； in, Represented as the corresponding data in the child posture data, , The sequence corresponds to the sitting pressure data, distance data, pitch angle, yaw angle, roll angle, neck and upper back angle, upper back and lower back angle, shoulder angle and elbow bending angle in the child posture data. is the reasonable interval of the corresponding data, It is represented as the normalized result of the corresponding data in the children's posture data; Finally, the normalized data ; AS3: Set weights for the corresponding data in the child posture data to indicate their influence on the rationality of the posture. The weights are set as follows: Sitting pressure data weight , distance data weight , the weight of joint angle data is , and the sum of the joint angle data weights is ; AS4: Establish a posture scoring model, and then use the following scoring function to reflect the posture scoring model; Sitting pressure data rationality score : Determine ideal sitting posture pressure data based on children's body shape , determine the score as: ； Distance data plausibility score :The ideal distance data is determined by the child's body shape , determine the score as: ； Joint angle data rationality score : Set the angle data of each joint The reasonable range is , determine the ideal angle based on the child's body shape , determine the score as: ； Comprehensive joint angle data rationality score : ； Then the overall rationality score is obtained by the following formula: : ； In the formula, , and are weight coefficients respectively.

4. A method for adjusting children's learning posture based on active learning according to claim 3, characterized in that: In step 3, the specific method of classifying the learning posture status of the child according to the overall rationality score to obtain the standard learning posture status, the good learning posture status, the learning posture status that needs to be improved, and the unreasonable learning posture status is: If the overall plausibility score , indicating the standard state of learning posture; If the overall plausibility score , indicating a good learning posture; If the overall plausibility score , indicating that the learning posture needs to be improved; If the overall plausibility score , indicating that the learning posture is unreasonable.

5. A method for adjusting children's learning posture based on active learning according to claim 4, characterized in that: In the step 4, the dynamic correction analysis is performed based on the previously acquired historical posture data of the child, and whether the current posture behavior is the posture habit of the child is determined according to the analysis result; and the specific content of determining whether to trigger the alarm information to prompt the child according to the judgment result is: BS1: Obtain children’s historical posture data and record children’s historical posture data within a specific time window: ； in, Indicates time The historical posture data of children at all times, including: historical sitting posture pressure data , historical distance data ;Historical joint angle data ; is the length of a specific time window; BS2: Statistical analysis of children’s historical posture data to build a mean model for children : ； BS3: Next, calculate the variance model of the child’s historical posture data : ； BS4: The child's posture data after obtaining the overall rationality score at the current moment is: ； in, The child's posture data collected at the current moment, including: current sitting posture pressure data , Current distance data ; Current joint angle data ; Calculate the current child posture data and the mean model of the child The deviation of : ； Then determine the normalized deviation vector by the following formula : ； Quantifying overall deviation using Euclidean distance : ； In the formula, represents the normalized bias vector The Euclidean norm of is calculated by summing the squares of each component of the deviation vector and taking the square root; Each component in Indicates the deviation size of a single feature; represents the normalized bias vector The kth component of ; The 9 in the figure refers to the 9 children’s posture data collected, including: sitting pressure data; distance data; pitch angle, yaw angle, roll angle, angle between neck and upper back, angle between upper back and lower back, shoulder angle and elbow bending angle; Set dynamic thresholds based on the statistical characteristics of children's historical posture data : ； In the formula, is an empirical parameter used to balance the detection of normal fluctuations and abnormal postures; Expressed as The overall deviation of the moment.

6. A method for adjusting children's learning posture based on active learning according to claim 5, characterized in that: The specific content of determining whether to trigger the alarm information to remind the child also includes: To make a correction judgment: when : Indicates that the current learning posture is the child’s posture habit, and the current learning posture is judged to be reasonable; when : Indicates that the current learning posture is not the child’s posture habit and triggers an alarm prompt directly.

7. A system for adjusting children's learning posture based on active learning, applied to a method for adjusting children's learning posture based on active learning according to any one of claims 1 to 6, characterized in that: include: A data collection module is used to collect children's posture data through a variety of sensors and cameras when the children are in a learning state; A posture assessment module, used for assessing the children's learning posture according to the children's posture data, and determining an overall rationality score; The posture judgment module is used to classify the learning posture status of children according to the overall rationality score, and obtain the learning posture standard status, the learning posture good status, the learning posture needs to be improved status and the learning posture unreasonable status; The intelligent correction module is used to perform dynamic correction analysis based on the previously acquired historical posture data of the child when the learning posture state is a state where the learning posture needs to be improved or a state where the learning posture is unreasonable, and to determine whether the current learning posture is the posture habit of the child according to the analysis result; and to determine whether to trigger an alarm message to remind the child according to the judgment result; The warning module is used to send out a warning message to prompt the child to adjust the posture when the intelligent correction module determines that the alarm information is triggered to prompt the child.

Images