Myopia-preventing and eye-protecting posture monitoring method and system

Abstract

The invention relates to the technical field of myopia protection, in particular to a posture monitoring method and system for preventing myopia and protecting eyes. During application, image data of a user is collected, and dynamic spatial positions of key nodes such as head eyes, neck point locations and trunk shoulders are extracted and processed. In a single second preset time period, the deviation intensity of the head posture, the neck posture and the trunk posture relative to the respective standard posture in the second preset time period is monitored, prompt information of the corresponding intensity is generated, a user can be reminded to correct unconscious posture deformation in time, the prompt intensity is matched with the deviation intensity, and the user experience is improved. And a prompting effect can be achieved, and the user cannot be excessively interfered, so that myopia inducements such as eyeball ciliary muscle fatigue, crystalline lens adjustment spasm and binocular adjustment imbalance caused by posture abnormity are avoided. By monitoring the deviation frequency of the three postures in the third preset time period, when the frequency exceeds a safety value, forced intervention information is generated, a dual protection mechanism is formed, and the myopia prevention effect is further enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of myopia prevention, in particular to a posture monitoring method and system for preventing myopia and protecting eyes. BACKGROUND

[0002] In daily scenes such as home learning and online course learning of teenagers and children, myopia problem is getting more and more serious, and bad learning posture is one of the key factors inducing myopia. When the head posture of a teenager or a child deviates from the standard during learning, the distance between the eyes and the viewing object such as books exceeds the safe range, which will cause the ciliary muscle of the eyeball to contract continuously, induce the accommodation spasm of the lens, and easily cause the elongation of the eye axis and accelerate the development of myopia. If the neck posture is abnormal and the lateral bending angle is too large, the head will be tilted, which will cause the horizontal height difference between the two eyes to increase, and the lens needs to be additionally adjusted to achieve focusing, which will accelerate the fatigue of the ciliary muscle and further increase the risk of myopia. Deviation of the trunk posture from the standard will also have adverse effects. When the trunk inclination angle is too large, the head will be difficult to self-adjust to the state of directly facing the front, which is easy to cause strabismus, so that the viewing distances of the two eyes are different, causing the imbalance of the two eyes, and then possibly leading to monocular myopia. During learning, these bad postures of the head, neck and trunk of teenagers or children are often formed unconsciously, and may gradually intensify during the entire learning process. If they cannot be discovered and corrected in time, they will cause continuous damage to the vision health. Therefore, how to monitor the body posture of a user to provide myopia prevention prompt information is a technical problem to be solved in the field. SUMMARY

[0003] In view of this, the present application provides a posture monitoring method and system for preventing myopia and protecting eyes, which can monitor the body posture of a user to provide myopia prevention prompt information.

[0004] In a first aspect, the present application provides a posture monitoring method for preventing myopia and protecting eyes, which is applied to a desk. The method comprises: starting to collect posture data of a user after a first preset time period after the user is seated; obtaining a head posture, a neck posture and a trunk posture according to the posture data; in a single learning stage, sequentially aggregating the posture data in a second preset time period; monitoring a first deviation intensity of the head posture relative to a standard head posture in the second preset time period, and generating first prompt information corresponding to the intensity; monitoring a second deviation intensity of the neck posture relative to a standard neck posture in the second preset time period, and generating second prompt information corresponding to the intensity; monitoring a third deviation intensity of the trunk posture relative to a standard trunk posture in the second preset time period, and generating third prompt information corresponding to the intensity; monitoring a deviation frequency of the head posture, the neck posture and the trunk posture in a third preset time period, which is greater than a preset threshold; and if the deviation frequency is greater than a safety value, generating forced intervention information.

[0005] With reference to the first aspect, in a possible implementation manner, the method further includes: obtaining a dynamic parameter according to the deviation frequency; and adjusting a length of the second preset time period of the next learning stage according to the dynamic parameter.

[0006] With reference to the first aspect, in a possible implementation manner, the obtaining the head posture, the neck posture and the trunk posture according to the posture data includes: extracting binocular image information from the posture data; and converting a binocular center position according to the binocular image information; the monitoring the first deviation intensity of the head posture relative to the head standard posture within the second preset time period, and generating the first prompt information corresponding to the intensity includes: monitoring a vertical relative distance and a horizontal relative distance of the binocular center position relative to a viewing reference point within the second preset time period; counting a first deviation number of times that the vertical relative distance deviates from a standard vertical distance by more than a vertical distance threshold; counting a second deviation number of times that the horizontal relative distance deviates from a standard horizontal distance by more than a horizontal distance threshold; obtaining a binocular deviation frequency according to the first deviation number of times and the second deviation number of times; and obtaining the first deviation intensity corresponding to the intensity according to the binocular deviation frequency, and obtaining the first prompt information corresponding to the intensity according to the first deviation intensity; wherein the greater the binocular deviation frequency is, the greater the first deviation intensity is.

[0007] With reference to the first aspect, in a possible implementation manner, the obtaining the head posture, the neck posture and the trunk posture according to the posture data includes: extracting neck image information from the posture data; extracting at least two neck feature points from the neck image information along a selected horizontal direction; and converting a neck lateral bending angle according to a plurality of spatial positions corresponding to the plurality of neck feature points; the monitoring the second deviation intensity of the neck posture relative to the neck standard posture within the second preset time period, and generating the second prompt information corresponding to the intensity includes: counting a neck deviation frequency that the neck lateral bending angle deviates from a standard lateral bending angle by more than a lateral bending threshold; and obtaining the second deviation intensity corresponding to the intensity according to the neck deviation frequency, and obtaining the second prompt information corresponding to the intensity according to the second deviation intensity; wherein the greater the neck deviation frequency is, the greater the second deviation intensity is.

[0008] With reference to the first aspect, in a possible implementation manner, after the extracting the at least two neck feature points from the neck image information, the method further includes: obtaining a real-time feature point spatial position corresponding to each of the neck feature points; monitoring the real-time feature point spatial position to obtain a first displacement speed corresponding to the neck feature point; and if a preset number of the first displacement speeds is greater than a first preset speed, skipping and not counting the neck deviation frequency.

[0009] In conjunction with the first aspect, in one possible implementation, after extracting binocular image information from the posture data, the method further includes: obtaining real-time eye spatial positions based on the binocular image information; monitoring the real-time eye spatial positions to obtain second displacement velocities of the eyes; and if the second displacement velocity of either eye is greater than a second preset velocity, then the neck deviation frequency is not included in the calculation.

[0010] In conjunction with the first aspect, in one possible implementation, obtaining the head posture, neck posture, and torso posture based on the posture data includes: extracting shoulder image information from the posture data; extracting two shoulder feature points located on the two shoulders respectively along a selected horizontal direction from the shoulder image information; and calculating the torso tilt angle based on multiple spatial positions corresponding to the two shoulder feature points; monitoring the third deviation intensity of the torso posture relative to the standard torso posture within a second preset time period and generating a third prompt message with corresponding intensity includes: counting the frequency of torso deviations where the torso tilt angle deviates from the standard tilt angle by more than a tilt threshold; and obtaining the corresponding third deviation intensity based on the torso deviation frequency, and obtaining the corresponding third prompt message based on the third deviation intensity; wherein, the greater the torso deviation frequency, the greater the third deviation intensity.

[0011] In conjunction with the first aspect, in one possible implementation, the side of the desk is provided with a first indicator light, a second indicator light, and a third indicator light sequentially along a first direction, where the first direction is the direction from the front of the user toward the user; the method further includes: generating a first control parameter based on the first indicator information, the first control parameter being used to control the first indicator light to flash; the flashing frequency and flashing brightness corresponding to the first control parameter are both proportional to the first deviation intensity; generating a second control parameter based on the second indicator information, the second control parameter being used to control the second indicator light to flash; the flashing frequency and flashing brightness corresponding to the second control parameter are both proportional to the second deviation intensity; and generating a third control parameter based on the third indicator information, the third control parameter being used to control the third indicator light to flash; the flashing frequency and flashing brightness corresponding to the third control parameter are both proportional to the third deviation intensity.

[0012] In conjunction with the first aspect, in one possible implementation, after generating mandatory intervention information if the deviation frequency is greater than the safety value, the method further includes: generating a fourth control parameter based on the mandatory intervention information, the fourth control parameter being used to control the first indicator light, the second indicator light, and the third indicator light to be constantly on or flashing together.

[0013] Secondly, this application provides a posture monitoring system for preventing myopia and protecting eyesight, applied to a desk; the system includes: an image acquisition module configured to acquire user posture data after a first preset time period following the user's seating; a data processing module communicatively connected to the image acquisition module, configured to obtain head posture, neck posture, and torso posture based on the posture data; and to summarize the posture data sequentially according to a second preset time period during a single learning phase; and a monitoring module communicatively connected to the data processing module, configured to monitor the first deviation intensity of the head posture from the standard head posture within the second preset time period, and generate a corresponding... The system includes: a first prompt message indicating the intensity of the deviation; monitoring the second deviation intensity of the neck posture relative to the standard neck posture within a second preset time period, and generating a second prompt message of corresponding intensity; monitoring the third deviation intensity of the torso posture relative to the standard torso posture within a second preset time period, and generating a third prompt message of corresponding intensity; monitoring the frequency of deviations of the head posture, neck posture, and torso posture exceeding a preset threshold within a third preset time period; the third preset time period being within a single learning phase; and a forced intervention module, communicatively connected to the monitoring module, configured to generate forced intervention information if the deviation frequency exceeds a safety value.

[0014] In application, this application effectively avoids invalid data in the initial unstable phase by delaying the collection of posture data for a first preset time period after the user sits down, thus enhancing the reference value of the posture data. By collecting the user's image data and extracting and processing the dynamic spatial positions of key nodes such as the head and eyes, neck points, and torso and shoulders, accurate dynamic posture data of the head, neck, and torso can be obtained. Individual learning stages are divided into second preset time periods, and posture data is aggregated to avoid the problem of poor reference value of single sets of data, making it suitable for learning scenarios of teenagers and children. Within each second preset time period, the deviation intensity of the head, neck, and torso posture from their respective standard postures is monitored, and corresponding prompts are generated. This promptly reminds the user to correct unconscious posture deformations, and the prompt intensity is matched to the deviation intensity, serving a prompting function without excessively interfering with the user's learning, thereby avoiding myopia-inducing factors such as ciliary muscle fatigue, lens spasm, and binocular accommodative imbalance caused by abnormal posture. Meanwhile, by monitoring the frequency of deviations in the three types of postures within a third preset time period, mandatory intervention information is generated when the frequency exceeds the safety value, forming a dual protection mechanism to further enhance the myopia prevention effect. Attached Figure Description

[0015] Figure 1 The diagram shows the steps of a posture monitoring method for preventing myopia and protecting eyesight, according to an embodiment of this application.

[0016] Figure 2 The diagram shows the steps for dynamically adjusting the second preset time period.

[0017] Figure 3 The diagram shows the specific steps of head monitoring.

[0018] Figure 4 The diagram shows the specific steps of neck monitoring.

[0019] Figure 5 The diagram shows the extraction of features from the neck and shoulders.

[0020] Figure 6 The diagram shows the handling of sudden changes in neck movement.

[0021] Figure 7 The diagram shows the processing of sudden changes in head movement.

[0022] Figure 8 The diagram shows the specific steps involved in trunk monitoring.

[0023] Figure 9 The diagram shows the steps involved in generating a prompt message for controlling the lights.

[0024] Figure 10 The diagram shows the steps involved in incorporating a mandatory intervention strategy.

[0025] Figure 11 The diagram shows the system structure of a posture monitoring system for myopia prevention and eye protection. Detailed Implementation

[0026] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0027] An exemplary posture monitoring method for preventing myopia and protecting eyesight is as follows:

[0028] Figure 1 The diagram illustrates the steps of a posture monitoring method for preventing myopia and protecting eyesight, according to an embodiment of this application. This application provides a posture monitoring method for preventing myopia and protecting eyesight, applied to a desk. A chair is provided in front of the desk. After the user sits down, the book to be viewed is located within a defined area of ​​the desk. Protrusions can be provided around the defined area to confine the book within it. In one embodiment, as... Figure 1 As shown, the posture monitoring method for preventing myopia and protecting eyesight includes:

[0029] Step 110: Start collecting the user's posture data after the first preset time period following the user's seating.

[0030] In this step, collecting posture data begins only after the user sits down, delaying the start of the initial, unstable posture data period immediately after sitting down. This avoids capturing posture data from the initial, unstable phase, making the collected posture data more valuable. Specifically, at least one depth-sensing camera is placed on the desk to capture the user's image. Correspondingly, step 110 and subsequent steps cease once the user leaves the desk.

[0031] Step 120: Obtain head posture, neck posture, and torso posture based on posture data.

[0032] In this step, existing motion capture systems (such as Vicon, OptiTrack, etc.) can be used for posture monitoring, and the captured image data is processed and transformed. Key human body nodes are marked in the image data. The key human body nodes of interest in this application include: the eyes of the head, the neck, and the shoulders of the torso. The dynamic spatial positions of the eyes are extracted from the posture data as dynamic head posture data; the dynamic spatial positions of at least two points on the neck are extracted from the posture data to calculate the lateral bending angle of the neck as dynamic neck posture data; and the dynamic spatial positions of the shoulders are extracted from the posture data as dynamic torso posture data.

[0033] Step 130: In a single learning phase, summarize the posture data sequentially according to the second preset time period.

[0034] In this step, using posture data spanning a continuous time range within the second preset time period as a reference avoids poor reference value due to small time spans in a single data set. Specifically, this application is mainly used for home learning and online learning for teenagers and children. The duration of a single learning session in this type of learning scenario is generally between 40 and 60 minutes. A single learning session can be divided into 2 to 6 time periods, that is, 2 to 4 sets of posture data are summarized sequentially within a single learning session. For example, if a single learning session is 60 minutes long and the number of segments is 6, then the second preset time period within a single learning session is 10 minutes. This means that each summary period of posture data is performed every 10 minutes, i.e., every 10 minutes after the start of a single learning session is a summary phase, and data is summarized sequentially for each second preset time period within a single learning session. The number and length of the second preset time period can be flexibly configured according to the intensity of myopia prevention monitoring, adapting to the intervention needs of different scenarios such as families and schools, effectively reducing the risk of myopia caused by poor learning posture in teenagers and children. The higher the monitoring intensity, the more second preset time periods there are in a single learning stage, and the shorter the duration of the second preset time period.

[0035] Step 140: Monitor the first deviation intensity of the head posture from the standard head posture within a second preset time period, and generate a first prompt message corresponding to the intensity.

[0036] In this step, users may unconsciously deform their movements during the learning process, leading to excessive deviations in head posture from the standard head posture. Timely monitoring and prompting are necessary. Within a single second preset time period, the deviation of the head posture from the standard head posture is monitored. A deviation exceeding a preset threshold is counted as a single head deviation. These single head deviations are accumulated within the single second preset time period, with different first deviation intensities corresponding to different accumulation counts. A corresponding first prompt is generated based on the first deviation intensity corresponding to the head posture, ensuring the first prompt is appropriately matched to the first deviation intensity to avoid being too strong or too weak. When the first deviation intensity is small, a weak prompt intensity can provide a noticeable effect without excessively impacting the user. Specifically, the first deviation intensity can be adjusted according to needs and application scenarios. For example, in head posture monitoring in some learning scenarios, the deviation intensity can be increased by one level every 5, 10, or 15 single head deviations. This number can be set as needed; this application is only an example. During the learning process, head posture has the most direct impact on the viewing distance between the eyes and the object being viewed. In other words, head posture directly affects whether the user's viewing distance is within a safe range, which corresponds to the standard head posture.

[0037] In general myopia prevention data, the standard vertical distance between the center of both eyes and the object being viewed is 30cm-40cm, and the standard horizontal distance is also 30cm-40cm. Excessive deviation from these standard distances can cause sustained contraction of the ciliary muscle, leading to accommodative spasm of the lens, which can eventually cause elongation of the eye axis and accelerate myopia development. In actual monitoring, the real-time vertical and horizontal relative distances between the center of both eyes are monitored to determine the initial deviation intensity.

[0038] Step 150: Monitor the second deviation intensity of the neck posture from the standard neck posture within a second preset time period, and generate a second prompt message corresponding to the intensity.

[0039] In this step, users may unconsciously deform their movements during the learning process, leading to excessive deviations in neck posture from the standard neck posture. Timely monitoring and prompting are necessary. Within a single second preset time period, the deviation of the neck posture from the standard neck posture is monitored. A deviation exceeding a preset threshold is counted as a single neck deviation. These single neck deviations are accumulated within the single second preset time period, with different deviation intensities corresponding to different accumulation counts. A corresponding second prompt is generated based on the second deviation intensity corresponding to the neck posture, ensuring the second prompt is appropriately matched to the second deviation intensity, avoiding overly strong or weak prompts. When the second deviation intensity is low, a weak prompt intensity can provide a noticeable effect without excessively impacting the user. Specifically, the second deviation intensity can be adjusted according to needs and application scenarios. For example, in some learning scenarios monitoring neck posture, the intensity can be increased by one for every 10 additional single deviations. This number can be set as needed; this application is only an example.

[0040] In general myopia prevention data, the standard lateral curvature angle of the neck is -3° to 3° relative to the vertical line along the neck's length. Within this range, the head can self-adjust to face forward, allowing both eyes to maintain a normal gaze, thus meeting myopia prevention standards. An excessively large curvature angle leads to head tilting, resulting in a greater difference in the horizontal height of the eyes. This requires the lens to adjust additionally to focus, accelerating ciliary muscle fatigue and thus accelerating myopia development. In actual monitoring, the real-time lateral curvature angle of the neck is monitored to obtain the second deviation intensity.

[0041] Step 160: Monitor the third deviation intensity of the torso posture from the standard torso posture within a second preset time period, and generate a third prompt message corresponding to the intensity.

[0042] In this step, users may unconsciously deform their movements during the learning process, leading to excessive deviations in torso posture from the standard torso posture. Timely monitoring and prompting are necessary. Within a single second preset time period, the deviation of the torso posture from the standard torso posture is monitored. If the deviation exceeds a preset threshold, it is counted as a single deviation. Within the single second preset time period, these single deviations are accumulated, with different third deviation intensities corresponding to different accumulation counts. A corresponding third prompt is obtained based on the third deviation intensity corresponding to the torso posture, ensuring that the third prompt is adapted to the third deviation intensity, avoiding overly strong or weak prompts. When the third deviation intensity is low, a weak prompt intensity can provide a noticeable effect without excessively affecting the user. Specifically, the third deviation intensity can be adjusted according to needs and application scenarios. For example, in some learning scenarios monitoring torso posture, the deviation intensity can be increased by one for every 10 additional single deviations. This number can be set as needed; this application is only an example.

[0043] In general myopia prevention data, the standard trunk posture can be set as follows: the perpendicular line from the line connecting the shoulders to the vertical plane should be -3° to 3° relative to the vertical line. Within this range, the head can self-adjust to keep it facing forward, thus maintaining normal vision and meeting myopia prevention standards. Excessive trunk tilt exceeds the head's self-adjustment capacity, leading to strabismus, which in turn causes different viewing distances for the two eyes, resulting in binocular accommodative imbalance and potentially causing unilateral myopia. In actual monitoring, the real-time trunk tilt angle is monitored to obtain the third deviation intensity.

[0044] Step 170: Monitor the frequency of deviations of head posture, neck posture and torso posture that are greater than a preset threshold within a third preset time period.

[0045] In this step, the third preset time period is within a single learning phase. Specifically, a time period within a single learning phase is selected as the third preset time period, and deviation frequencies are accumulated within this time period. Specifically, each deviation of the head posture from the standard head posture exceeding a preset threshold is counted as one deviation frequency; the same applies to the neck posture and torso posture. Within a single learning phase, one third preset time period can be configured, i.e., step 170 is executed once; multiple third preset time periods can also be configured, i.e., step 170 is executed multiple times. The number of third preset time periods can be configured based on the intensity of myopia prevention monitoring. Higher intensity monitoring indicates greater emphasis on myopia protection for the user, and consequently, more third preset time periods are available within a single learning phase. The intensity of myopia prevention monitoring can be set by the user, their guardian, their teacher, or the system. The number and length of the third preset time period can be flexibly configured according to the intensity of myopia prevention monitoring, adapting to the intervention needs of different scenarios such as families and schools, effectively reducing the risk of myopia caused by poor learning posture in teenagers and children. The higher the monitoring intensity, the more third preset time periods there are in a single learning stage, and the shorter the duration of the third preset time period.

[0046] Step 180: If the deviation frequency is greater than the safety value, stop executing step 170 and generate mandatory intervention information.

[0047] In steps 170 and 180, the second and third preset time periods can be calculated independently. Within the third preset time period, the total frequency of deviations in head posture, neck posture, and torso posture is counted. When the counted deviation frequency exceeds a safe value, it indicates frequent abnormalities in head posture, neck posture, and torso posture, which can exacerbate myopia and necessitates mandatory intervention. Specifically, the mandatory intervention information can be used to instruct the execution of at least one of the following actions: issuing a voice alarm, turning off lights, vibrating the desk, or moving the chair away from the desk, etc. If this application is applied to a multi-person classroom setting in a school, a vibrator can be installed in the desk to achieve a mandatory intervention effect and avoid disturbing others. If this application is applied to a single-person learning setting in a home, a voice alarm can be used to achieve a mandatory intervention effect. This step, through the safety value, provides dual protection and further enhances the myopia prevention effect. Specifically, the size of the third preset time period can be set according to the intensity of the monitored myopia prevention measures. A higher intensity of monitored myopia prevention measures indicates greater emphasis on myopia protection for the user; correspondingly, a higher intensity of monitored myopia prevention measures results in a shorter third preset time period. The intensity of myopia prevention monitoring can be set by the user, the user's guardian, the user's teacher, or the system. In some embodiments, the third preset time period can be calculated as a multiple of the second preset time period, thereby reducing the amount of computation. For example, the third preset time period can be 1, 1.5, 2, 2.5, or 3 times the second preset time period, etc.

[0048] In this embodiment, by delaying the collection of posture data for a first preset time period after the user sits down, invalid data in the initial unstable phase is effectively avoided, thus enhancing the reference value of the posture data. By collecting the user's image data and extracting and processing the dynamic spatial positions of key nodes such as the head and eyes, neck points, and torso and shoulders, accurate dynamic posture data of the head, neck, and torso can be obtained. Each learning stage is divided into a second preset time period, and the posture data is summarized, avoiding the problem of poor reference value of single sets of data, making it suitable for learning scenarios of teenagers and children. Within each second preset time period, the deviation intensity of the head, neck, and torso posture from their respective standard postures is monitored, and corresponding prompts are generated. This promptly reminds the user to correct unconscious posture deformations, and the prompt intensity is matched to the deviation intensity, serving a prompting function without excessively interfering with the user's learning, thereby avoiding myopia-inducing factors such as ciliary muscle fatigue, lens spasm, and binocular accommodative imbalance caused by abnormal posture. Meanwhile, by monitoring the frequency of deviations in the three types of postures within a third preset time period, mandatory intervention information is generated when the frequency exceeds the safety value, forming a dual protection mechanism to further enhance the myopia prevention effect.

[0049] Figure 2 The diagram illustrates the steps of a method for dynamically adjusting a second preset time period. In one embodiment, as shown...Figure 2 As shown, this posture monitoring method for preventing myopia and protecting eyesight also includes:

[0050] Step 190: Obtain dynamic parameters based on the deviation frequency.

[0051] Step 200: Adjust the duration of the second preset time period for the next learning stage according to the dynamic parameters.

[0052] This embodiment can dynamically adjust the duration of the second preset time period. The greater the frequency of deviation, the more unstable the user's state is, and the more the monitoring intensity of the user needs to be strengthened. That is, the duration of the second preset time period is reduced, thereby increasing the monitoring frequency of the user to better play the role of preventing myopia and protecting eyesight.

[0053] Specifically, dynamic parameters can be derived based on the safety value. When the deviation frequency equals the safety value, the dynamic parameter is 1; when the deviation frequency is more than twice the safety value, the dynamic parameter can be set to 2; when the deviation frequency is a multiple between 1 and 2 times the safety value, the dynamic parameter is equal to that multiple; when the deviation frequency is less than 0.5 times the safety value, the dynamic parameter can be set to 0.5; when the deviation frequency is a multiple between 0.5 and 1 times the safety value, the dynamic parameter is equal to that multiple. Then, the duration of the second preset time period of the next learning stage is obtained by dividing the current third preset time period by the dynamic parameter.

[0054] Figure 3 The diagram illustrates the specific steps of head monitoring. Specifically, in one embodiment, as shown... Figure 3 As shown, step 120 includes:

[0055] Step 1201: Extract binocular image information from pose data.

[0056] This application employs a depth-sensing camera device. The posture data captured by the depth-sensing camera device contains depth information. Then, based on binocular feature recognition, binocular image information corresponding to both eyes is extracted from the posture data.

[0057] Step 1202: Calculate the center position of both eyes based on the image information of both eyes.

[0058] In this step, the coordinates of the midpoint of the line connecting the two eyes are taken as the coordinates of the center of the two eyes from the binocular image information. The coordinates of the center of the two eyes obtained by the depth-sensing camera are essentially the spatial position of the center of the two eyes relative to the depth-sensing camera, that is, the position of the center of the two eyes in step 1202.

[0059] Step 140 includes:

[0060] In step 141, during the second preset time period, monitor the vertical and horizontal relative distances between the center positions of both eyes and the viewing reference point.

[0061] In this step, the viewing reference point is set as the geometric center of the object to be viewed. Since the depth-sensing camera is fixed on the desk and the position of the object to be viewed is fixed, based on the relative positions of the depth-sensing camera and the object to be viewed, the geometric center of the depth-sensing camera, the object to be viewed, and the centers of the eyes form a triangle. The vertical and horizontal relative distances between the centers of the eyes and the viewing reference point can then be calculated.

[0062] Step 142: Count the first number of deviations where the vertical relative distance deviates from the standard vertical distance and exceeds the vertical distance threshold.

[0063] In general myopia prevention data, the standard vertical distance between the center of both eyes and the object being viewed that meets myopia prevention standards is 30cm to 40cm. The vertical distance threshold can be set to 5cm, 10cm, or 15cm. When the difference between the vertical relative distance and 30cm to 40cm is greater than the vertical distance threshold, a deviation is counted, and the first deviation count is accumulated within a second preset time period. After the vertical relative distance returns to the standard vertical distance range, monitoring of the vertical relative distance begins again. When the vertical relative distance continues to deviate from the vertical distance threshold, the first deviation count is recorded once per second.

[0064] Step 143: Count the second number of deviations where the relative horizontal distance deviates from the standard horizontal distance beyond the horizontal distance threshold.

[0065] In general myopia prevention data, the standard horizontal distance between the center of both eyes and the object being viewed is 30cm to 40cm, and the horizontal distance threshold can be set to 5cm, 10cm, or 15cm. When the difference between the horizontal relative distance and 30cm to 40cm is greater than the horizontal distance threshold, a deviation is counted, and the second deviation count is accumulated over a second preset time period. After the horizontal relative distance returns to the standard horizontal distance range, monitoring of the horizontal relative distance begins again. When the horizontal relative distance continues to deviate from the horizontal distance threshold, the second deviation count is recorded once per second.

[0066] Step 144: Obtain the binocular deviation frequency based on the first and second deviation counts.

[0067] In this step, the first deviation count and the second deviation count are added together to obtain the binocular deviation frequency.

[0068] Step 145: Obtain the corresponding first deviation intensity based on the frequency of deviation between the two eyes, and obtain the corresponding first prompt information based on the first deviation intensity.

[0069] In this step, the greater the frequency of eye deviation, the greater the intensity of the first deviation, thus obtaining the corresponding first prompt message. Specifically, for example, it can be set that for every 5, 10, or 15 additional instances of eye deviation, the intensity of the first deviation increases by one level.

[0070] In this embodiment, the spatial coordinates of the center positions of the eyes are calculated by converting the image information of both eyes. Then, the vertical relative distance and horizontal relative distance are obtained from the center positions of the eyes. The frequency of deviations of the vertical relative distance and horizontal relative distance are accumulated respectively, and finally the deviation frequency of the eyes is obtained. The corresponding first prompt information is obtained based on the deviation frequency of the eyes.

[0071] Figure 4 The diagram shows the specific steps of neck monitoring. Figure 5 The diagram illustrates the extraction of neck and shoulder features. Specifically, in one embodiment, as shown... Figure 4 As shown, step 120 includes:

[0072] Step 1211: Extract neck image information from posture data.

[0073] This application employs a depth-sensing camera device. The posture data captured by the depth-sensing camera device contains depth information, and then the entire neck image information is extracted from the posture data based on neck feature recognition.

[0074] Step 1212: Extract at least two neck feature points from the neck image information along the selected horizontal direction.

[0075] In this step, refer to Figure 5 From the neck image information, several image points are randomly selected along a chosen horizontal direction L1 as neck feature points Q1. The spatial coordinates of the neck feature point Q1 are essentially the spatial position of the neck feature point Q1 relative to the depth-sensing camera device. The horizontal direction L1 can be set as a horizontal line passing through the midpoint of the neck.

[0076] Step 1213: Calculate the lateral bending angle of the neck based on the multiple spatial positions corresponding to multiple neck feature points.

[0077] In this step, multiple neck feature points Q1 are connected to each other, and a perpendicular line from this line to the vertical plane is drawn as the neck reference line S1. When the neck is bent laterally, the angle between the neck reference line S1 and the vertical line can be calculated to obtain the neck lateral bending angle.

[0078] Step 150 includes:

[0079] Step 151: Count the frequency of neck deviations where the lateral bending angle deviates from the standard lateral bending angle by more than the lateral bending threshold.

[0080] In common myopia prevention data, the standard lateral bending angle of the neck is -3° to 3° relative to the vertical line along the neck's length. The bending threshold can be set to 2°, 4°, or 6°. When the difference between the lateral bending angle and -3° to 3° is greater than the bending threshold, a deviation is counted. The number of neck deviations is accumulated over a second preset time period. After the neck returns to the standard lateral bending angle range, monitoring of the lateral bending angle resumes. When the neck continues to deviate beyond the bending threshold, the frequency of neck deviation is counted once per second.

[0081] Step 152: Obtain the corresponding second deviation intensity based on the frequency of neck deviation, and obtain the corresponding second prompt information based on the second deviation intensity.

[0082] In this step, the greater the frequency of neck deviation, the greater the intensity of the second deviation, thus obtaining the corresponding second prompt information. Specifically, for example, it can be set that for every 5, 10, or 15 additional instances of neck deviation, the intensity of the second deviation increases by one level.

[0083] In this embodiment, the spatial coordinates of neck feature points are calculated by converting neck image information, points are randomly selected and connected, and the neck lateral bending angle is obtained according to the rotation angle of the connecting line. The neck lateral bending angle is compared with the standard lateral bending angle, and finally the frequency of neck deviation is obtained. The corresponding second prompt information is obtained according to the frequency of neck deviation.

[0084] Figure 6 The diagram illustrates the processing of sudden changes in neck movement. In one embodiment, after step 1212, as... Figure 6 As shown, this posture monitoring method for preventing myopia and protecting eyesight also includes:

[0085] Step 1214: Obtain the corresponding real-time spatial location of each neck feature point.

[0086] Step 1215: Monitor the real-time spatial position of the feature points to obtain the first displacement velocity of the corresponding neck feature points.

[0087] Step 1216: If the first displacement velocity of the preset number is greater than the first preset velocity, then skip it and do not count it in the neck deviation frequency.

[0088] In this embodiment, the real-time spatial position of each neck feature point is monitored, and the corresponding first displacement velocity is also monitored. The preset number can be set to be greater than or equal to 2. When the first displacement velocity of the preset number of neck feature points is greater than the first preset velocity, it indicates that the head and neck are moving or rotating rapidly. Generally, the first preset velocity can be set to 0.5 m / s. A velocity greater than 0.5 m / s can be considered a sudden movement. This movement may mean that the user is moving their neck. Such situations can be considered sudden movements. These sudden movements are unrelated to learning and are not unconscious movement deformations. They do not need to be included in the frequency of neck deviation and can be directly ignored. This embodiment can avoid misjudgment caused by the above special cases.

[0089] Regarding the setting of the first preset speed, when a healthy user's neck rapidly bends and swings to the left and right, the peak angular velocity is approximately 200-400° / s. According to the linear velocity formula v=ω×r, where ω is the angular velocity and its unit is rad / s, 1° / s=π / 180≈0.01745rad / s. Based on the average person's neck size, the radius r of the neck's side-bending swing can be taken as 0.1m. Taking 200° / s as an example, when the neck bends rapidly to the side, the minimum linear velocity v=200*0.01745*0.1=0.349m / s. Therefore, setting the first preset speed to 0.3m / s is sufficient to cover most sudden changes in neck movement.

[0090] Figure 7 The diagram illustrates the processing of sudden changes in head movement. In one embodiment, after step 1201, as... Figure 7 As shown, this posture monitoring method for preventing myopia and protecting eyesight also includes:

[0091] Step 1203: Obtain the real-time spatial position of the eyes based on the binocular image information.

[0092] Step 1204: Monitor the real-time spatial position of the eye to obtain the second displacement velocity of the eye.

[0093] Step 1205: If the second displacement velocity of any eye is greater than the second preset velocity, then the neck deviation frequency is not counted.

[0094] In this embodiment, the spatial position of each eye is monitored in real time, and the second displacement velocity corresponding to the eye's spatial position is detected. When the second displacement velocity of any eye is greater than a second preset velocity, it indicates that the head is moving rapidly. Generally, the second preset velocity can be set to 0.3 m / s. A velocity greater than 0.3 m / s can be considered a sudden change in movement. This movement may mean that the user is moving their neck, looking at other places, or being called by the outside world. Such situations can be considered sudden changes in movement. These sudden changes in movement are unrelated to learning and are not unconscious movement deformations. They do not need to be included in the frequency of neck deviation and can be directly ignored. This embodiment can avoid misjudgment caused by the above-mentioned special situations.

[0095] Regarding the setting of the second preset speed, when a healthy user rapidly shakes their head left and right, the peak angular velocity is approximately 300-600° / s. The rotation axis of the eyes is roughly along the longitudinal axis of the cervical spine (near the midline), and the horizontal distance from the eyes to this axis is approximately 0.08-0.12 m. Taking 0.1 m as a reference value, the minimum linear velocity of the eyes during rapid head shaking is v = 300 * 0.01745 * 0.1 = 0.5235 m / s. When a healthy user rapidly swings their head left and right, the peak angular velocity is approximately 200-400° / s. The eyes swing left and right with the root of the jaw as the axis, and the horizontal distance from the eyes to the root of the jaw is approximately 0.08-0.12 m. Taking 0.1 m as a reference value, the minimum linear velocity of the eyes during rapid head shaking is v = 200 * 0.01745 * 0.1 = 0.349 m / s. In summary, setting the second preset speed to 0.3 m / s is sufficient to cover most sudden changes in head movement.

[0096] Figure 8 The diagram illustrates the specific steps of trunk monitoring. In one embodiment, as shown... Figure 8 As shown, step 120 includes:

[0097] Step 1221: Extract shoulder image information from posture data.

[0098] This application employs a depth-sensing camera device. The posture data captured by the depth-sensing camera device contains depth information. Then, based on shoulder feature recognition, the corresponding shoulder image information is extracted from the posture data.

[0099] Step 1222: Extract two shoulder feature points located on the two shoulders respectively from the shoulder image information along the selected horizontal direction.

[0100] Step 1223: Calculate the torso tilt angle based on the multiple spatial positions corresponding to the two shoulder feature points.

[0101] In this step, refer to Figure 5 Connect two shoulder feature points Q2 on the selected horizontal direction L2, and draw a perpendicular line from this line to the vertical plane as the shoulder reference line S2. When the torso is tilted, the angle between the shoulder reference line S2 and the vertical line can be calculated to obtain the torso tilt angle.

[0102] Step 160 includes:

[0103] Step 161: Count the frequency of trunk deviations where the trunk tilt angle deviates from the standard tilt angle by more than the tilt threshold.

[0104] In the general myopia prevention data, the standard inclination and lateral bending angles are as follows: the perpendicular line of the shoulder connection line in the vertical plane with respect to the vertical line is -3° to 3°. The inclination threshold can be set to 2°, 4°, or 6°. When the difference between the trunk inclination angle and -3° to 3° is greater than the inclination threshold, it is counted as one deviation, and the number of trunk deviations is accumulated within the second preset time period. After the trunk returns to the standard inclination angle range, the monitoring of the trunk inclination angle starts again. When the trunk continuously deviates beyond the inclination threshold, the trunk deviation frequency is counted once per second.

[0105] Step 162: Obtain the corresponding third deviation intensity according to the trunk deviation frequency, and obtain the corresponding third prompt information according to the third deviation intensity.

[0106] In this step, the greater the trunk deviation frequency, the greater the third deviation intensity, and thus the corresponding third prompt information is obtained. Specifically, for example, it can be set that for every additional 5, 10, or 15 times of the trunk deviation frequency, the third deviation intensity increases by one intensity level.

[0107] Figure 9 Shown is a schematic diagram of the method steps for generating the prompt information for controlling the lamp. In one embodiment, a first prompt lamp, a second prompt lamp, and a third prompt lamp are sequentially arranged on the side of the desk along the first direction. The first direction is the direction from in front of the user towards the user, that is, the third prompt lamp, the second prompt lamp, and the first prompt lamp are successively farther away from the user. As Figure 9 shown, the posture monitoring method for preventing myopia and protecting eyesight further includes:

[0108] Step 210: Generate a first control parameter according to the first prompt information, and the first control parameter is used to control the first prompt lamp to blink.

[0109] In this step, both the blinking frequency and the blinking brightness corresponding to the first control parameter are directly proportional to the first deviation intensity, that is, the greater the first deviation intensity, the faster the blinking frequency and the higher the blinking brightness of the first prompt lamp, that is, the greater the reminder intensity. The smaller the first deviation intensity, it means that the deformation of the user's head posture is not large, and the lower blinking frequency and blinking brightness can avoid excessive influence on the user. The attention priority of the head posture is relatively high, so the first prompt lamp is located in the frontmost position, making it the easiest for the user to notice. In some specific embodiments, an LED lamp in the shape of the character "head" can be made as the first prompt lamp, so as to intuitively prompt the user to pay attention to the head posture.

[0110] Step 220: Generate a second control parameter according to the second prompt information, and the second control parameter is used to control the second prompt lamp to blink.

[0111] In this step, both the flashing frequency and the flashing brightness corresponding to the second control parameter are directly proportional to the second deviation intensity. That is, the greater the second deviation intensity, the faster the flashing frequency and the higher the flashing brightness of the second indicator light, which means the greater the reminder intensity. The smaller the second deviation intensity, it indicates that the deformation of the user's neck posture is not significant. The lower flashing frequency and flashing brightness can avoid excessive influence on the user. The user can notice the second indicator light for prompting the neck posture through peripheral vision. In some specific embodiments, an LED light in the shape of the character "neck" can be made as the second indicator light, so as to visually prompt the user to pay attention to the neck posture.

[0112] Step 230: Generate a third control parameter according to the third prompt information, and the third control parameter is used to control the third indicator light to flash.

[0113] In this step, both the flashing frequency and the flashing brightness corresponding to the third control parameter are directly proportional to the third deviation intensity. That is, the greater the third deviation intensity, the faster the flashing frequency and the higher the flashing brightness of the third indicator light, which means the greater the reminder intensity. The smaller the third deviation intensity, it indicates that the deformation of the user's torso posture is not significant. The lower flashing frequency and flashing brightness can avoid excessive influence on the user. The third indicator light is located at the oblique side of the user, and the user can notice the third indicator light for prompting the torso posture through peripheral vision. This position can allow the user to intuitively feel that it corresponds to the user's torso rather than the head and neck. In some specific embodiments, an LED light in the shape of the character "body" can be made as the third indicator light, so as to visually prompt the user to pay attention to the torso posture.

[0114] Figure 10 The following shows the schematic diagram of the method steps after adding the forced intervention strategy. In one embodiment, as Figure 10 shown, after step 180, the posture monitoring method for preventing myopia and protecting eyesight further includes:

[0115] Step 240: Generate a fourth control parameter according to the forced intervention information, and the fourth control parameter is used to control the first indicator light, the second indicator light, and the third indicator light to light up or flash together.

[0116] In this embodiment, when forced intervention is required, a fourth control parameter is generated, which can be used to instruct the first indicator light, the second indicator light, and the third indicator light to light up or flash together, so as to effectively achieve the effect of forced intervention.

[0117] An exemplary posture monitoring system for preventing myopia and protecting eyesight is as follows:

[0118] Figure 11 The following shows the schematic diagram of the system structure of the posture monitoring system for preventing myopia and protecting eyesight. The present application provides a posture monitoring system for preventing myopia and protecting eyesight, which is applied to a desk, as Figure 11As shown, the system includes: an image acquisition module 1101, a data processing module 1102, a monitoring module 1103, and a forced intervention module 1104.

[0119] The image acquisition module 1101 is configured to start acquiring the user's posture data after the first preset time period following the user's seating.

[0120] The data processing module 1102 is communicatively connected to the image acquisition module 1101. The data processing module 1102 is configured to: obtain head posture, neck posture, and torso posture based on posture data; and summarize posture data sequentially according to a second preset time period during a single learning phase.

[0121] The monitoring module 1103 is communicatively connected to the data processing module 1102. The monitoring module 1103 is configured to: monitor the first deviation intensity of the head posture relative to the standard head posture within a second preset time period, and generate a first prompt message corresponding to the intensity; monitor the second deviation intensity of the neck posture relative to the standard neck posture within a second preset time period, and generate a second prompt message corresponding to the intensity; monitor the third deviation intensity of the torso posture relative to the standard torso posture within a second preset time period, and generate a third prompt message corresponding to the intensity; monitor the frequency of deviations of the head posture, neck posture, and torso posture exceeding a preset threshold within a third preset time period; and the third preset time period is in a single learning phase.

[0122] The mandatory intervention module 1104 is communicatively connected to the monitoring module 1103. The mandatory intervention module 1104 is configured to generate mandatory intervention information if the deviation frequency is greater than the safety value.

[0123] In this embodiment, by delaying the collection of posture data for a first preset time period after the user sits down, invalid data in the initial unstable phase is effectively avoided, thus enhancing the reference value of the posture data. By collecting the user's image data and extracting and processing the dynamic spatial positions of key nodes such as the head and eyes, neck points, and torso and shoulders, accurate dynamic posture data of the head, neck, and torso can be obtained. Each learning stage is divided into a second preset time period, and the posture data is summarized, avoiding the problem of poor reference value of single sets of data, making it suitable for learning scenarios of teenagers and children. Within each second preset time period, the deviation intensity of the head, neck, and torso posture from their respective standard postures is monitored, and corresponding prompts are generated. This promptly reminds the user to correct unconscious posture deformations, and the prompt intensity is matched to the deviation intensity, serving a prompting function without excessively interfering with the user's learning, thereby avoiding myopia-inducing factors such as ciliary muscle fatigue, lens spasm, and binocular accommodative imbalance caused by abnormal posture. Meanwhile, by monitoring the frequency of deviations in the three types of postures within a third preset time period, mandatory intervention information is generated when the frequency exceeds the safety value, forming a dual protection mechanism to further enhance the myopia prevention effect.

[0124] The basic principles of this application have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the application to the necessity of employing the aforementioned specific details for implementation.

[0125] The block diagrams of devices, apparatuses, devices, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0126] It should also be noted that in the apparatus, equipment, and methods of this application, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of this application.

[0127] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this application. Therefore, this application is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features of the invention herein.

[0128] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications or equivalent substitutions made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A posture monitoring method for preventing myopia and protecting eyesight, characterized in that, Applied to a desk; the method includes: The user's posture data will be collected starting one preset time period after the user sits down. The head posture, neck posture, and torso posture are obtained based on the posture data. In a single learning phase, the posture data is summarized sequentially according to the second preset time period. Monitor the intensity of the first deviation of the head posture from the standard head posture within a second preset time period, and generate a first prompt message with the corresponding intensity. Monitor the intensity of the second deviation of the neck posture from the standard neck posture within a second preset time period, and generate a second prompt message with the corresponding intensity. Monitor the intensity of the third deviation of the torso posture from the standard torso posture within a second preset time period, and generate a third prompt message of corresponding intensity; The frequency of deviations of the head posture, neck posture, and torso posture exceeding a preset threshold within a third preset time period is monitored; the third preset time period is within a single learning phase; and If the frequency of deviations exceeds the safety value, mandatory intervention information is generated.

2. The posture monitoring method for preventing myopia and protecting eyesight according to claim 1, characterized in that, Also includes: Dynamic parameters are obtained based on the deviation frequency; as well as The duration of the second preset time period in the next learning phase is adjusted according to dynamic parameters.

3. The posture monitoring method for preventing myopia and protecting eyesight according to claim 1, characterized in that, The process of obtaining head posture, neck posture, and torso posture based on the posture data includes: Extract binocular image information from the posture data; and Calculate the center position of both eyes based on the binocular image information; The monitoring of the first deviation intensity of the head posture from the standard head posture within a second preset time period, and the generation of a first prompt message corresponding to the intensity, includes: During the second preset time period, monitor the vertical and horizontal relative distances between the center positions of the two eyes and the viewing reference point; Count the number of first deviations where the vertical relative distance deviates from the standard vertical distance by more than the vertical distance threshold; Count the number of second deviations where the horizontal relative distance deviates from the standard horizontal distance by more than the horizontal distance threshold; The binocular deviation frequency is obtained based on the first deviation count and the second deviation count; and The first deviation intensity is obtained based on the frequency of deviation of both eyes, and the first prompt information is obtained based on the first deviation intensity; wherein, the greater the frequency of deviation of both eyes, the greater the first deviation intensity.

4. The posture monitoring method for preventing myopia and protecting eyesight according to claim 3, characterized in that, The process of obtaining head posture, neck posture, and torso posture based on the posture data includes: Extract neck image information from the posture data; From the neck image information, at least two neck feature points are extracted along a selected horizontal direction; and The lateral bending angle of the neck is calculated based on the multiple spatial positions corresponding to the multiple neck feature points. The monitoring of the second deviation intensity of the neck posture from the standard neck posture within a second preset time period, and the generation of a corresponding second prompt message, includes: Count the frequency of neck deviations where the neck lateral curvature angle deviates from the standard lateral curvature angle by more than a lateral curvature threshold; and The second deviation intensity is obtained based on the frequency of neck deviation, and the second prompt information is obtained based on the second deviation intensity; wherein, the greater the frequency of neck deviation, the greater the second deviation intensity.

5. The posture monitoring method for preventing myopia and protecting eyesight according to claim 4, characterized in that, After extracting at least two neck feature points from the neck image information, the method further includes: The corresponding real-time spatial position of the feature point is obtained based on each of the neck feature points; The first displacement velocity corresponding to the neck feature point is obtained by monitoring the real-time spatial position of the feature point. If the first displacement velocity of the preset number is greater than the first preset velocity, then it is skipped and not included in the neck deviation frequency.

6. The posture monitoring method for preventing myopia and protecting eyesight according to claim 4, characterized in that, After extracting binocular image information from the pose data, the method further includes: The real-time spatial position of the eyes is obtained based on the binocular image information; The second displacement velocity of the eye is obtained by monitoring the real-time spatial position of the eye; If the second displacement velocity of any eye is greater than the second preset velocity, then the neck deviation frequency is not included in the calculation.

7. The posture monitoring method for preventing myopia and protecting eyesight according to claim 1, characterized in that, The process of obtaining head posture, neck posture, and torso posture based on the posture data includes: Extract shoulder image information from the posture data; From the shoulder image information, extract two shoulder feature points located at the two shoulders respectively along a selected horizontal direction; and The torso tilt angle is calculated based on the multiple spatial positions corresponding to the two shoulder feature points. The monitoring of the third deviation intensity of the torso posture from the standard torso posture within a second preset time period, and the generation of a third prompt message corresponding to the intensity, includes: Count the frequency of trunk deviations where the trunk tilt angle deviates from the standard tilt angle by more than a tilt threshold; and The third deviation intensity is obtained based on the frequency of trunk deviation, and the third prompt information is obtained based on the third deviation intensity; wherein, the greater the frequency of trunk deviation, the greater the third deviation intensity.

8. The posture monitoring method for preventing myopia and protecting eyesight according to claim 1, characterized in that, The side of the desk is provided with a first indicator light, a second indicator light and a third indicator light in sequence along a first direction, the first direction being the direction from the front of the user toward the user; The method further includes: A first control parameter is generated based on the first prompt information. The first control parameter is used to control the first prompt light to flash. The flashing frequency and flashing brightness corresponding to the first control parameter are both proportional to the first deviation intensity. A second control parameter is generated based on the second prompt information. This second control parameter controls the second prompt light to flash. The flashing frequency and brightness of the second control parameter are both proportional to the second deviation intensity. A third control parameter is generated based on the third prompt information. The third control parameter is used to control the third prompt light to flash. The flashing frequency and flashing brightness corresponding to the third control parameter are both proportional to the third deviation intensity.

9. The posture monitoring method for preventing myopia and protecting eyesight according to claim 8, characterized in that, After generating mandatory intervention information if the deviation frequency exceeds a safety value, the method further includes: A fourth control parameter is generated based on the forced intervention information. The fourth control parameter is used to control the first indicator light, the second indicator light, and the third indicator light to be constantly on or flashing together.

10. A posture monitoring system for preventing myopia and protecting eyesight, characterized in that, Applied to a desk; the system includes: The image acquisition module is configured to start acquiring the user's posture data after the first preset time period following the user's seating. The data processing module is communicatively connected to the image acquisition module. The data processing module is configured to: obtain head posture, neck posture, and torso posture based on the posture data; and summarize the posture data sequentially according to a second preset time period during a single learning phase. A monitoring module, communicatively connected to the data processing module, is configured to: monitor the first deviation intensity of the head posture relative to the standard head posture within a second preset time period, and generate a first prompt message corresponding to the corresponding intensity; monitor the second deviation intensity of the neck posture relative to the standard neck posture within a second preset time period, and generate a second prompt message corresponding to the corresponding intensity; monitor the third deviation intensity of the torso posture relative to the standard torso posture within a second preset time period, and generate a third prompt message corresponding to the corresponding intensity; monitor the frequency of deviations of the head posture, neck posture, and torso posture exceeding a preset threshold within a third preset time period; wherein the third preset time period is within a single learning phase; and A mandatory intervention module is communicatively connected to the monitoring module. The mandatory intervention module is configured to generate mandatory intervention information if the deviation frequency is greater than the safety value.

Images