Intelligent health monitoring and service system assisted by artificial intelligence

Abstract

The invention relates to the technical field of health service, and particularly discloses an intelligent health monitoring and service system assisted by artificial intelligence, which is used for solving the problem that the existing intelligent eye protection equipment is difficult to correct accurately when the posture of a user changes. Parameters influencing eyesight detection are obtained through the first detection unit, the second detection unit and the third parameter obtaining unit, correction needing to be conducted during eyesight detection is conducted based on parameter analysis, the problem that a traditional eyesight protection instrument is difficult to correct accurately when the posture of a user changes is solved through real-time detection and dynamic adjustment, and the user experience is improved. The accuracy and effectiveness of eyesight protection are remarkably improved, and the user experience and the health management effect are enhanced through an intelligent and personalized correction scheme.

Description

Technical Field

[0001] The present invention relates to the field of health service technology, and more specifically, to an artificial intelligence-assisted intelligent health monitoring and service system. Background Art

[0002] For example, a Chinese invention patent with publication number CN107622645A discloses an intelligent device for vision protection, which is used to assist vision protection when learning to read and doing homework, including: a light environment sensor, a distance sensor, a central processing unit, a display prompt module, an operation button, a communication module, and a power supply module; the light environment sensor measures the illumination, color temperature, and stroboscopic parameters of the location of the object being read, and determines through the central processing unit whether the health indicators are met, and gives a prompt if not; the distance sensor measures the distance from the object being read to the human eye, and determines through the central processing unit whether it is lower than the minimum distance requirement, and gives a prompt if it is lower; the user's cumulative reading volume is calculated based on the illumination and distance measurement results, and determines whether it exceeds the threshold, and gives a prompt if it exceeds the threshold, so that the light environment health can be detected, and the user can be assisted in maintaining a correct sitting posture and avoiding watching electronic screens for a long time, thereby preventing myopia and protecting eyesight.

[0003] The above patents all have the problem raised by this background technology: the existing vision protection smart devices are difficult to implement when using distance sensors to measure distance and correct eye habits, and can only judge whether the user's reading distance is qualified based on the distance measurement results. It is worth noting that if the user is not facing the distance sensor, the vision protector will be difficult to correct accurately, but the user cannot always maintain the same posture, which will lead to deviations in the correction. In order to solve the above problems, a technical solution is now provided. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides an artificial intelligence-assisted intelligent health monitoring and service system, which obtains parameters affecting vision detection through a first detection unit, a second detection unit and a third parameter acquisition unit, and performs corrections required for vision detection based on parameter analysis. Through real-time detection and dynamic adjustment, the problem that existing intelligent eye protection devices are difficult to accurately correct when the user's posture changes is solved, thereby solving the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: An artificial intelligence-assisted intelligent health monitoring and service system comprises a vision protector, an intelligent vision recognition management module and a vision correction adjustment module; the vision protector is connected to the intelligent vision recognition management module, and the intelligent vision recognition management module is connected to the vision correction adjustment module; the intelligent vision recognition management module comprises a camera and a first detection unit; the camera is used to capture a user's facial image in real time and obtain a user's standard facial image; the first detection unit is used to extract the user's standard facial image and segment the user's standard facial image into a number of facial key areas, set key points in each facial key area, determine the standard eye area and the position parameters of the key points in the standard eye area; then extract the user's facial image and perform a detection on the user's facial image; The facial image is segmented into several facial key areas, key points are set in each facial key area, and position parameters of the eye area and the key points in the eye area are determined; the vision correction adjustment module includes a first deviation adjustment module; the first deviation adjustment module is used to extract the standard eye area and the position parameters of the key points in the standard eye area in the first detection unit and the real-time eye area and the position parameters of the key points in the eye area; the number of standard key points is obtained based on the standard eye area, and the number of real-time key points is obtained based on the real-time eye area; the position parameters of the key points in the standard eye area and the position parameters of the key points in the real-time eye area are imported into the first deviation analysis model to calculate the first deviation value, and the formula of the first deviation analysis model is: ; Where: is the first deviation value, is the total time that the user spends reading. is the number of standard key points, is the number of key points in the real-time eye area at the i-th moment, is the number of key points in the real-time eye area at the i+1th moment, is the mean of the sequence of key points in the set.

[0006] As a further solution of the present invention, the mean of the concentrated key point number sequence is obtained by obtaining the real-time eye area and the position parameters of the key points in the eye area, obtaining the real-time key point number according to the real-time eye area of ​​the user at each moment during the reading time, and arranging the key point number in the real-time eye area at each moment in order from small to large to obtain the key point number sequence, removing the maximum value and the minimum value in the key point number sequence to obtain the concentrated key point number sequence, and calculating the formula for the mean of the concentrated key point number sequence is: ; Where: is the mean of the sequence of key points in the set, is the number of key points in the real-time eye area at the i-th moment in the sequence of concentrated key points, The total number of moments the user spends reading.

[0007] As a further solution of the present invention, the intelligent vision recognition management module also includes a cloud data processing center, a second detection unit and a third parameter acquisition unit; the vision correction adjustment module also includes a second deviation adjustment module, a third deviation adjustment module and a vision detection correction module.

[0008] As a further solution of the present invention, the second detection unit is used to obtain the head feature change value of the user during the reading time; the head feature change value is obtained by obtaining the head movement speed change value of the user at each moment during the reading time, and the head movement speed change value is obtained by the change of the position parameters of the key points in the eye area of ​​the user at each moment during the reading time. The specific calculation formula for obtaining is: ; Where: is the head feature change value at time i, is the position coordinate of the key point h in the eye area at time i+1, is the position coordinate of the key point h in the eye area at time i, is the time interval between time i+1 and time i.

[0009] As a further solution of the present invention, the third parameter acquisition unit is used to perform ultrasonic ranging through the eyesight protector to obtain the first ranging length range parameters between each key point in the user's standard facial image and the eyesight protector, and to obtain the second ranging length range parameters between the facial key points at each moment during the user's reading time and the eyesight protector.

[0010] As a further solution of the present invention, the second deviation adjustment module is used to extract the head feature change value during the user's reading time, and to construct a second deviation analysis model based on the head feature change value to calculate the second deviation value. The formula of the second deviation analysis model is: ; Where: is the second deviation value, is the total time that the user spends reading. is the position coordinate of the key point h in the eye area at time i+2, is the time interval between time i+2 and time i+1, is the position coordinate of the key point h in the eye area at time i+1, is the position coordinate of the key point h in the eye area at time i, is the time interval between time i+1 and time i.

[0011] As a further solution of the present invention, the third deviation adjustment module is used to extract the first ranging length range parameter and the second ranging length range parameter, import the first ranging length range parameter and the second ranging length range parameter into the second deviation analysis model, and calculate the third deviation value. The formula of the third deviation analysis model is: ; Where: is the third deviation value, is the first distance measurement length parameter between the jth key point in the user's standard facial image and the vision protection device, is the mean value of the first distance measurement length parameter between each key point in the user's standard facial image and the vision protection device, is the first maximum distance measurement length parameter between each key point in the user's standard facial image and the vision protection device, is the first minimum distance measurement length parameter between each key point in the user's standard facial image and the vision protection device, is the number of key points in the user's standard facial image, To prevent the adjustment parameter from having a zero denominator, is the number of facial key points at time i, is the second distance measurement length parameter between the e-th facial key point and the vision protection device at time i, is the mean value of the second distance measurement length parameter between the facial key point and the vision protection device at time i, is the second maximum distance measurement parameter between the facial key point and the vision protection device at time i, is the second minimum distance measurement length parameter between the facial key point and the vision protection device at time i, The total number of moments the user spends reading.

[0012] As a further solution of the present invention, the vision detection correction module analyzes the vision detection correction value based on the first deviation value, the second deviation value and the third deviation value, obtains the preset standard reading distance, imports the first deviation value, the second deviation value, the third deviation value and the preset standard reading distance into the vision detection correction model, and calculates the vision detection correction value. The formula of the vision detection correction model is: ; Where: is the visual acuity test correction value, is the adjustment factor, is the reading distance output by the vision protection device at time i, is the preset standard reading distance. is the third deviation value, is the second deviation value, is the first deviation value.

[0013] The technical effects and advantages of the artificial intelligence-assisted intelligent health monitoring and service system of the present invention: the present invention obtains parameters that affect vision detection through a first detection unit, a second detection unit and a third parameter acquisition unit, and correction is required when performing vision detection based on parameter analysis. Through real-time detection and dynamic adjustment, the problem that existing intelligent eye protection devices are difficult to accurately correct when the user's posture changes is solved. Not only does it significantly improve the accuracy and effectiveness of vision protection, but it also enhances the user experience and health management effects through intelligent and personalized correction solutions, ultimately helping users to better protect their eyesight and prevent vision problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A schematic diagram of the structure of an artificial intelligence-assisted intelligent health monitoring and service system provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0015] The following will be combined with the accompanying drawings in the present invention to clearly and completely describe the technical solution in the present invention. Obviously, the described technical solution is only a part of the present invention, not all of it. Based on the technical solution in the present invention, all other technical solutions obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0016] Figure 1 FIG. 1 shows a schematic diagram of the structure of an artificial intelligence-assisted intelligent health monitoring and service system provided by Embodiment 1 of the present invention. Figure 1 As shown, an artificial intelligence-assisted intelligent health monitoring and service system in this embodiment includes a vision protector, an intelligent vision recognition management module and a vision correction adjustment module, the vision protector is connected to the intelligent vision recognition management module, and the intelligent vision recognition management module is connected to the vision correction adjustment module; The vision protector is used to measure the distance between the user's facial image and the vision protector by ultrasonic wave; The intelligent vision recognition management module is used to obtain parameters that affect vision detection; The vision correction analysis and adjustment module is used to analyze the corrections required when performing vision testing using a vision protection device.

[0017] The intelligent vision recognition management module includes a cloud data processing center, a camera, a first detection unit, a second detection unit, and a third parameter acquisition unit; the camera is connected to the first detection unit and the second detection unit respectively, and the first detection unit, the second detection unit, and the third parameter acquisition unit are connected to the cloud data processing center respectively; The cloud data processing center is used to store the parameters acquired by the first detection unit, the second detection unit and the third parameter acquisition unit; The camera is used to capture the user's facial image in real time and obtain the user's standard facial image; The first detection unit is used to extract the user's standard facial image, segment the user's standard facial image into a number of facial key areas, set key points in each facial key area, and determine the standard eye area and the position parameters of the key points in the standard eye area; then extract the user's facial image, segment the user's facial image into a number of facial key areas, set key points in each facial key area, and determine the eye area and the position parameters of the key points in the eye area; The second detection unit is used to obtain the head feature change value of the user during the reading time; the head feature change value is obtained by obtaining the head movement speed change value of the user at each moment during the reading time, and the head movement speed change value is obtained by the change of the position parameters of the key points in the eye area of ​​the user at each moment during the reading time. The specific calculation formula for obtaining is: ; Where: is the head feature change value at time i, is the position coordinate of the key point h in the eye area at time i+1, is the position coordinate of the key point h in the eye area at time i, is the time interval between time i+1 and time i.

[0018] The third parameter acquisition unit is used to perform ultrasonic ranging through the eyesight protector to obtain the first ranging length range parameters between each key point in the user's standard facial image and the eyesight protector, and to obtain the second ranging length range parameters between the facial key points at each moment during the user's reading time and the eyesight protector.

[0019] The camera can capture the user's facial image in real time and immediately obtain the user's standard facial image. This enables the system to obtain the latest data at any time when the user is performing vision-related activities, ensuring the timeliness and accuracy of the information; the first detection unit performs detailed segmentation of the user's facial image and sets key points in key areas, especially determining the position parameters of the standard eye area and key points in the eye area, which can more accurately evaluate and adjust the user's vision protection measures; through the calculation of the head feature change value of the second detection unit, the user's head movement state can be monitored in real time, and the vision protection measures can be dynamically adjusted according to these data to provide immediate feedback. Different types of parameters are collected through multiple detection units and combined with these parameters for comprehensive analysis, such as using the head feature change value and the ranging length range parameter to correct and optimize vision detection, which can improve the detection accuracy and correction effect of the system; based on the real-time collected data, the system can automatically adjust the vision correction strategy, such as: when it is detected that the user's head movement amplitude is too large, the system can issue a warning in time or automatically adjust the vision protection measures, thereby avoiding vision problems caused by bad posture.

[0020] The vision correction adjustment module includes a first deviation adjustment module, a second deviation adjustment module, a third deviation adjustment module and a vision detection correction module; the first deviation adjustment module, the second deviation adjustment module and the third deviation adjustment module are respectively connected to the vision detection correction module; The first deviation adjustment module is used to extract the position parameters of the standard eye area and the key points in the standard eye area in the first detection unit and the position parameters of the real-time eye area and the key points in the eye area; obtain the number of standard key points based on the standard eye area and obtain the number of real-time key points based on the real-time eye area; import the position parameters of the key points in the standard eye area and the position parameters of the key points in the real-time eye area into the first deviation analysis model to calculate the first deviation value, and the formula of the first deviation analysis model is: ; Where: is the first deviation value, is the total time that the user spends reading. is the number of standard key points, is the number of key points in the real-time eye area at the i-th moment, is the number of key points in the real-time eye area at the i+1th moment, is the mean of the sequence of key points in the set.

[0021] During the user's reading process, the changes in the number of key points in the eye area are dynamically monitored. By comparing the standard eye area and the current eye area in real time, the deviation of the user's posture or perspective can be quickly detected, thereby providing timely feedback and adjustment suggestions; by comparing the number of standard key points and real-time key points, and combining the sequence mean of the number of key points, the changes in the eye area can be analyzed more accurately, which helps to improve the accuracy of vision correction and prevent vision deviation caused by slight changes in posture or angle; the deviation is quantified by calculating the change in the number of real-time eye key points, so that the deviation analysis no longer relies on a single measurement value, but is evaluated by the average change at multiple times. The evaluation method based on quantity changes is more reasonable and stable, and avoids the impact of accidental errors on the overall correction; through continuous deviation adjustment, it helps users maintain a relatively stable and comfortable reading posture, thereby reducing visual fatigue caused by improper posture, helping to improve the user's reading experience, and protecting the user's vision health.

[0022] The real-time eye area and the position parameters of the key points in the eye area are obtained. According to the real-time eye area of ​​the user at each moment during the reading time, the number of real-time key points is obtained, and the number of key points in the real-time eye area at each moment is arranged in order from small to large to obtain a key point number sequence. The maximum and minimum values ​​in the key point number sequence are removed to obtain a concentrated key point number sequence. The formula for calculating the mean of the concentrated key point number sequence is: ; Where: is the mean of the sequence of key points in the concentration, is the number of key points in the real-time eye area at the i-th moment in the sequence of concentrated key points, The total number of moments the user spends reading.

[0023] By removing the maximum and minimum values ​​in the key point quantity sequence, outliers can be effectively filtered out. These outliers may be caused by short-term measurement errors or sudden head movements or eye closures of the user. After removing these abnormal data, the calculated mean can better represent the user's actual eye status; the concentrated key point quantity sequence excludes extreme values, making the calculated mean more stable, improving the accuracy of the system's response to real-time changes in the user's eye area, and avoiding inaccurate overall deviation analysis results due to a few abnormal situations; by calculating the mean of the key point quantity sequence after removing extreme values, a more stable indicator is obtained as the basic data for judging changes in the user's eye area, which can more accurately reflect the user's eye status most of the time, thereby improving the accuracy of deviation analysis.

[0024] The second deviation adjustment module is used to extract the head feature change value during the user's reading time, and build a second deviation analysis model based on the head feature change value to calculate the second deviation value. The formula of the second deviation analysis model is: ; Where: is the second deviation value, is the total time that the user spends reading. is the position coordinate of the key point h in the eye area at time i+2, is the time interval between time i+2 and time i+1, is the position coordinate of the key point h in the eye area at time i+1, is the position coordinate of the key point h in the eye area at time i, is the time interval between time i+1 and time i.

[0025] in, It represents the movement speed of the eye key point h between time i+1 and i+2; It represents the movement speed of the eye key point h between time i and i+1. The second deviation value is obtained by accumulating the difference between the two speed values ​​and finally taking the average of these differences, which reflects the stability or fluctuation of the user's eye position during the reading process. A larger deviation value means that the user's head or eye position has a larger change or instability during reading.

[0026] By calculating the movement speed of key eye points at different time points and comparing the changes in these speeds, the stability of the user's head and eye positions can be detected very accurately, which helps to identify tiny head or eye movements, so as to promptly discover posture changes that may affect vision. When the system detects a larger second deviation value, it means that the user's head or eye position has changed significantly during the reading process. The calculation and analysis of the deviation value enables the system to dynamically adjust vision protection measures according to the user's actual eye use. For detected unstable posture or excessive head movement, it can automatically correct or prompt the user, thereby improving the accuracy and effectiveness of vision protection. Since each user's reading posture and habits may be different, this module can perform personalized analysis and adjustment based on the user's specific head movement pattern, which improves the flexibility of the system and enables it to better meet the needs of different users. By detecting and feeding back the stability of the head and eyes, it can help users maintain a good reading posture, reduce visual fatigue caused by head or eye instability, help improve the user's reading experience, and make long-term eye use more comfortable.

[0027] The third deviation adjustment module is used to extract the first ranging length range parameter and the second ranging length range parameter, import the first ranging length range parameter and the second ranging length range parameter into the second deviation analysis model, and calculate the third deviation value. The formula of the third deviation analysis model is: ; Where: is the third deviation value, is the first distance measurement length parameter between the jth key point in the user's standard facial image and the vision protection device, is the mean value of the first distance measurement length parameter between each key point in the user's standard facial image and the vision protection device, is the first maximum distance measurement length parameter between each key point in the user's standard facial image and the vision protection device, is the first minimum distance measurement length parameter between each key point in the user's standard facial image and the vision protection device, is the number of key points in the user's standard facial image, To prevent the adjustment parameter from having a zero denominator, is the number of facial key points at time i, is the second distance measurement length parameter between the e-th facial key point and the vision protection device at time i, is the mean value of the second distance measurement length parameter between the facial key point and the vision protection device at time i, is the second maximum distance measurement parameter between the facial key point and the vision protection device at time i, is the second minimum distance measurement length parameter between the facial key point and the vision protection device at time i, The total number of moments the user spends reading.

[0028] in, It represents the accumulation of the normalized absolute differences between all facial key points and the distance measurement mean at each moment during the entire reading time. It is used to measure the distance measurement changes between facial key points. It expresses the sum of the normalized differences of the distance measurement parameters of the facial key points at each moment during the user's reading period relative to the moment mean, indicating the distance measurement stability during the entire reading time. It represents the average value of the normalized absolute difference between all key points in the user's standard facial image and the ranging mean, providing a standard reference for comparison with the ranging situation at time i. It expresses the average value of the normalized difference of the ranging parameters of all key points in the standard facial image relative to the mean, indicating a stable ranging benchmark.

[0029] By accumulating the distance measurement data at each moment during the entire reading time, it is possible to comprehensively consider the changes in the user's head position during the entire reading process, capture all subtle changes that may affect vision, and thus provide a more comprehensive analysis; through normalization processing, it is possible to effectively eliminate the impact of differences in the magnitude of different distance measurement data, making the data at different moments and between different users comparable, and making the deviation analysis more fair and accurate, avoiding the deviation in the distance measurement results caused by individual differences or device characteristics; by calculating the distance measurement mean and the average value of the normalized difference of the user's standard facial image, the system establishes a stable benchmark, allowing the system to determine whether the user's head position deviates from the initial standard position in different time periods, so as to promptly discover and correct bad eye behavior; the third deviation adjustment module can accurately monitor the changes in the user's head position during the reading process, and compare it with the standard benchmark to provide personalized vision protection adjustments, which not only improves the accuracy and reliability of the system, but also enhances the user experience and helps to effectively protect vision.

[0030] The vision detection correction module analyzes the vision detection correction value based on the first deviation value, the second deviation value, and the third deviation value, obtains a preset standard reading distance, imports the first deviation value, the second deviation value, the third deviation value, and the preset standard reading distance into the vision detection correction model, and calculates the vision detection correction value. The formula of the vision detection correction model is: ; Where: is the visual acuity test correction value, is the adjustment factor, is the reading distance output by the vision protection device at time i, is the preset standard reading distance. is the third deviation value, is the second deviation value, is the first deviation value.

[0031] Compare the vision detection correction value with the preset vision detection correction threshold. If the vision detection correction value is greater than or equal to the preset vision detection correction threshold, then there is a deviation in the vision correction of the user by the vision protector; if the vision detection correction value is less than the preset vision detection correction threshold, then there is no deviation in the vision correction of the user by the vision protector.

[0032] By comprehensively analyzing the first deviation value, the second deviation value and the third deviation value, combining them with the preset standard reading distance, and calculating the vision detection correction value, the multi-dimensional analysis method can more comprehensively reflect the various deviations that may exist in the actual reading process of the user, thereby performing more accurate vision correction; by introducing the adjustment coefficient, the influence of each deviation value on the vision correction value can be flexibly adjusted according to the specific situation, so that the correction model can adapt to the needs of different users and different usage environments, thereby improving the accuracy of vision correction; by calculating the vision detection correction value in real time and comparing it with the preset threshold, the deviation of the user's vision correction by the vision protector can be discovered in time, and adjustments or prompts can be made to the user when necessary, thereby improving the effectiveness of vision protection.

[0033] The adjustment coefficient is determined by the historical data of the cloud data processing center. An initial adjustment coefficient is set by extracting the reading distance output by the vision protection device at time i in the historical data, the preset standard reading distance, the third deviation value, the second deviation value and the first deviation value. , substituted into the vision test correction model formula for calculation, and the difference between the model prediction value and the target correction value is measured by defining the error function, where the error function formula is defined as: ; Where: is the error function output result, is the number of samples, is the predicted visual acuity test correction value of group c samples, is the target visual acuity test correction value for group c samples; Then the error function is minimized by the gradient descent optimization algorithm. The formula of the gradient descent optimization algorithm is: ; Where: is the updated adjustment coefficient, is the adjustment coefficient before updating, is the learning rate, is the derivative of the error function with respect to the adjustment coefficient, For time; The coefficients are adjusted by continuously updating the iteration until the output of the error function converges to the minimum value or reaches the preset error threshold.

[0034] The embodiment of the present invention obtains parameters that affect vision detection through a first detection unit, a second detection unit and a third parameter acquisition unit, and performs corrections required for vision detection based on parameter analysis. Through real-time detection and dynamic adjustment, it solves the problem that existing intelligent eye protection devices are difficult to accurately correct when the user's posture changes. It not only significantly improves the accuracy and effectiveness of vision protection, but also enhances user experience and health management effects through intelligent and personalized correction solutions, ultimately helping users to better protect their eyesight and prevent vision problems.

[0035] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0036] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An artificial intelligence-assisted intelligent health monitoring and service system, comprising a vision protector, an intelligent vision recognition management module and a vision correction adjustment module; the vision protector is connected to the intelligent vision recognition management module, and the intelligent vision recognition management module is connected to the vision correction adjustment module; characterized in that, The intelligent vision recognition management module includes a camera and a first detection unit; the camera is used to capture the user's facial image in real time and obtain the user's standard facial image; the first detection unit is used to extract the user's standard facial image and segment the user's standard facial image into a number of facial key areas, set key points in each facial key area, and determine the standard eye area and the position parameters of the key points in the standard eye area; then extract the user's facial image and segment the user's facial image into a number of facial key areas, set key points in each facial key area, and determine the eye area and the position parameters of the key points in the eye area; the vision correction adjustment module includes a first deviation adjustment module; The first deviation adjustment module is used to extract the position parameters of the standard eye area and key points in the standard eye area in the first detection unit and the position parameters of the real-time eye area and key points in the eye area; Obtain the number of standard key points based on the standard eye area, and obtain the number of real-time key points based on the real-time eye area; The position parameters of the key points in the standard eye area and the position parameters of the key points in the real-time eye area are imported into the first deviation analysis model to calculate the first deviation value. The formula of the first deviation analysis model is: ; Where: is the first deviation value, is the total time that the user spends reading. is the number of standard key points, is the number of key points in the real-time eye area at the i-th moment, is the number of key points in the real-time eye area at the i+1th moment, is the mean of the sequence of key points in the set.

2. The artificial intelligence-assisted intelligent health monitoring and service system according to claim 1, characterized in that: The mean of the concentrated key point number sequence is obtained by obtaining the real-time eye area and the position parameters of the key points in the eye area, obtaining the number of real-time key points according to the real-time eye area of ​​the user at each moment during the reading time, and arranging the number of key points in the real-time eye area at each moment in order from small to large to obtain the key point number sequence, removing the maximum and minimum values ​​in the key point number sequence to obtain the concentrated key point number sequence, and calculating the formula for the mean of the concentrated key point number sequence is: ; Where: is the mean of the sequence of key points in the set, is the number of key points in the real-time eye area at the i-th moment in the sequence of concentrated key points, The total number of moments the user spends reading.

3. The artificial intelligence-assisted intelligent health monitoring and service system according to claim 1, characterized in that: The intelligent vision recognition management module also includes a cloud data processing center, a second detection unit and a third parameter acquisition unit; the vision correction adjustment module also includes a second deviation adjustment module, a third deviation adjustment module and a vision detection correction module.

4. The artificial intelligence-assisted intelligent health monitoring and service system according to claim 3, characterized in that: The second detection unit is used to obtain the head feature change value of the user during the reading time; The head feature change value is obtained by obtaining the change value of the head movement speed of the user at each moment during the reading time. The head movement speed change value is obtained by the change of the position parameters of the key points in the eye area of ​​the user at each moment during the reading time. The specific calculation formula is: ; Where: is the head feature change value at time i, is the position coordinate of the key point h in the eye area at time i+1, is the position coordinate of the key point h in the eye area at time i, is the time interval between time i+1 and time i.

5. The artificial intelligence-assisted intelligent health monitoring and service system according to claim 3, characterized in that: The third parameter acquisition unit is used to perform ultrasonic ranging through the eyesight protector to obtain the first ranging length range parameters between each key point in the user's standard facial image and the eyesight protector, and to obtain the second ranging length range parameters between the facial key points at each moment during the user's reading time and the eyesight protector.

6. The artificial intelligence-assisted intelligent health monitoring and service system according to claim 3, characterized in that: The second deviation adjustment module is used to extract the head feature change value during the user's reading time, and build a second deviation analysis model based on the head feature change value to calculate the second deviation value. The formula of the second deviation analysis model is: ; Where: is the second deviation value, is the total time that the user spends reading. is the position coordinate of the key point h in the eye area at time i+2, is the time interval between time i+2 and time i+1, is the position coordinate of the key point h in the eye area at time i+1, is the position coordinate of the key point h in the eye area at time i, is the time interval between time i+1 and time i.

7. The artificial intelligence-assisted intelligent health monitoring and service system according to claim 3, characterized in that: The third deviation adjustment module is used to extract the first ranging length range parameter and the second ranging length range parameter, import the first ranging length range parameter and the second ranging length range parameter into the second deviation analysis model, and calculate the third deviation value. The formula of the third deviation analysis model is: ; Where: is the third deviation value, is the first distance measurement length parameter between the jth key point in the user's standard facial image and the vision protection device, is the mean value of the first distance measurement length parameter between each key point in the user's standard facial image and the vision protection device, is the first maximum distance measurement length parameter between each key point in the user's standard facial image and the vision protection device, is the first minimum distance measurement length parameter between each key point in the user's standard facial image and the vision protection device, is the number of key points in the user's standard facial image, To prevent the adjustment parameter from having a zero denominator, is the number of facial key points at time i, is the second distance measurement length parameter between the e-th facial key point and the vision protection device at time i, is the mean value of the second distance measurement length parameter between the facial key point and the vision protection device at time i, is the second maximum distance measurement parameter between the facial key point and the vision protection device at time i, is the second minimum distance measurement length parameter between the facial key point and the vision protection device at time i, The total number of moments the user spends reading.

8. The artificial intelligence-assisted intelligent health monitoring and service system according to claim 3, characterized in that: The vision detection correction module analyzes the vision detection correction value based on the first deviation value, the second deviation value, and the third deviation value, obtains a preset standard reading distance, imports the first deviation value, the second deviation value, the third deviation value, and the preset standard reading distance into the vision detection correction model, and calculates the vision detection correction value. The formula of the vision detection correction model is: ； Where: is the visual acuity test correction value, is the adjustment factor, is the reading distance output by the vision protection device at time i, is the preset standard reading distance. is the third deviation value, is the second deviation value, is the first deviation value.

Images