Office state recognition and healthy office promotion method

Abstract

The invention discloses an office state recognition and healthy office promotion method. The method comprises the following steps: information collection and feature point extraction; office scene judgment; analyzing sitting posture, standing posture and eye features; an office health identification state; the bad health state is reminded and improved; and performing dynamic monitoring and identification analysis. According to the method, multi-dimensional quantitative features are comprehensively considered, sitting posture features and standing posture features of an office scene are considered, eye features are also considered, and the features are graded and quantized; and through comprehensive analysis of the features, the healthy office state of the user can be more accurately judged, so that the accuracy and reliability of health monitoring are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent office, and in particular relates to a method for identifying office status and promoting healthy office. Background Art

[0002] With the development of science and technology, human society has entered an information and digital age. In modern society, office work has become an indispensable part of people's daily life. However, long-term work in front of the computer has a great impact on people's physical health, which may lead to vision loss and may also cause office occupational diseases such as cervical spondylosis and lumbar spondylosis. Therefore, how to effectively achieve a healthy working state has become an important research topic.

[0003] In the field of healthy office status recognition: Existing technologies mainly collect user sitting posture information through sensors or cameras to determine whether the user is in a healthy office state. There is a problem that the single indicator cannot effectively identify other office status characteristics. With the rapid development of the lifting table industry, standing office has become a trend. Similar technologies cannot identify the health status in standing office scenes; secondly, there are some technologies that identify the length of sitting and standing postures, and only judge the healthy office status from the perspective of long-term sitting or long-term standing.

[0004] In the field of healthy office management: Existing technologies mainly solve the problem of long-term sedentary work by sending reminders to users, adjusting sitting postures, and alternating between sitting and standing, but ignore the attention paid to eye health in office scenes. In addition, existing technologies only judge healthy office status in two states: healthy and unhealthy. No more in-depth graded and quantitative research has been conducted, and it is impossible to provide more precise health promotion plans.

[0005] Therefore, how to build a method that comprehensively utilizes the sitting posture, standing posture and eye features in the office scene to identify the level of the user's office status and quantify the severity, while providing health promotion suggestions and adjustment measures for poor office status is an important challenge facing current technology. Summary of the invention

[0006] In order to make up for the deficiencies of the prior art, the present invention aims to provide a method for office status recognition and healthy office promotion, which comprehensively considers multi-dimensional quantitative characteristics to improve the accuracy and reliability of health monitoring.

[0007] The technical problem to be solved by the present invention can be achieved through the following technical solutions:

[0008] The method for identifying office status and promoting healthy office work comprises the following steps:

[0009] S1. Information collection and feature point extraction;

[0010] S2. Office scene judgment;

[0011] S3. Analysis of sitting posture, standing posture and eye features;

[0012] S4. Office health identification status;

[0013] S5. Reminder and improvement of adverse health conditions;

[0014] S6 dynamic monitoring and identification analysis.

[0015] Furthermore, the step S1 includes the following contents:

[0016] Collecting user office scene information using a scene information collection device: collecting office sitting posture information using an office chair pressure sensor device, collecting office standing posture information using a head and shoulder position collection device, and collecting eye information using an eye feature collection device;

[0017] Extract feature points of the user's office scene, sitting posture, standing posture, and eyes.

[0018] Furthermore, in step S2, a preset office scene judgment rule is used to judge the user's office scene based on the scene information, sitting posture information, standing posture information and eye information; wherein, the office scene judgment rule is as follows: if the angle between the facial orientation and the gaze direction is within the threshold range of the screen gaze, it is judged as staring at the screen. Under this premise, if the screen gaze time reaches the threshold, the scene is judged to be an office scene; if the screen gaze time does not reach the threshold, the scene is judged to be a non-office scene.

[0019] Furthermore, step S3 includes the following contents:

[0020] Calculate scene information based on the user's office scene feature points; the scene information includes the user's facial orientation, gaze direction, and duration of looking at the screen;

[0021] Calculate the sitting posture information according to the user's office sitting posture feature points; the sitting posture information includes leaning forward, hunching over, half lying down, and crossing legs;

[0022] Calculate the standing posture information according to the user's office standing posture feature points; the standing posture information includes leaning forward, bending over and hunching over;

[0023] Eye information is calculated based on the user's eye feature points; the eye information includes the distance from the eye to the screen, blinking frequency, conjunctival congestion, and dull eyes.

[0024] Furthermore, in the step S4, the user's office health status is determined based on the office scene information, sitting posture information, standing posture information and eye information using a preset office health status determination rule;

[0025] The office health status judgment rule is as follows: according to the preset healthy office status, it is divided into very healthy, healthy, general, unhealthy, and very unhealthy, and corresponds to the corresponding threshold range. According to the calculated total score of the sitting posture information, standing posture information, and eye information, the healthy office status can be judged into five states: very healthy, healthy, general, unhealthy, and very unhealthy.

[0026] When the total feature score is judged as very healthy, healthy or general eye health status, if 1 or 2 feature levels reach the maximum threshold, it is judged as an unhealthy working state; if 3 or more feature levels reach the maximum threshold, it is judged as a very unhealthy working state;

[0027] When two eye features reach the maximum threshold, it is considered an unhealthy working state; when three or more eye features reach the maximum threshold, it is considered a very unhealthy working state.

[0028] Furthermore, in step S5, the preset office health status adjustment rules are used to remind the user to make adjustments based on the office scene information, sitting posture information, standing posture information and eye information, or actively adjust the height / working mode of the office desk and chair to improve the user's bad office habits and allow the user to have a healthy office state.

[0029] Furthermore, the user office scene is obtained by:

[0030] Obtain facial feature points in office scenes;

[0031] Extract facial orientation, gaze direction, screen gaze start time, and screen gaze end time;

[0032] According to the facial orientation and the gaze direction, the angle between the facial orientation and the gaze direction is calculated to determine whether the user is looking at the screen;

[0033] The screen gaze duration is calculated according to the start time and the end time of screen gaze;

[0034] The information obtained in the above steps can be used as a basis for judging, monitoring, and adjusting whether the office is healthy in subsequent steps.

[0035] Furthermore, the sitting posture information can be obtained in the following manner:

[0036] Get sitting posture feature points;

[0037] Extract the pressure data of each coordinate point based on the sitting posture feature points;

[0038] Based on the pressure data of the coordinate points, determine whether the sitting posture is leaning forward, hunched over, half-lying, or crossing legs, and the degree of each sitting posture feature, marked as 0-5 (0-5 is the degree value, 0 means that the feature does not appear, and 1-5 increases and becomes more and more serious, the same below);

[0039] The above information can be used as a basis for judging, monitoring and adjusting whether office work is healthy in subsequent steps.

[0040] Furthermore, the standing posture information can be obtained by the following calculation method:

[0041] Get standing posture feature points;

[0042] Extract the coordinate points of shoulders, pupils and other parts based on the standing posture feature points;

[0043] Based on the coordinate points, judge whether the standing posture is leaning forward, hunched over, and the degree of each standing posture feature, marked as 0-5;

[0044] The above information can be used as a basis for judging, monitoring and adjusting whether office work is healthy in subsequent steps.

[0045] Furthermore, the eye information can be obtained by:

[0046] Obtain eye image feature points;

[0047] Based on the feature points of the eye image, the distance from the eye to the screen, the blinking frequency, and the quantization level of these eye features are calculated, marked as 0-5;

[0048] Based on the feature points of the eye image, determine whether the eye has conjunctival congestion, dull eyesight, and the degree of the feature, marked as 0-5;

[0049] The above information can be used as a basis for judging, monitoring and adjusting whether office work is healthy in subsequent steps.

[0050] Compared with the prior art, the present invention has the following advantages:

[0051] (1) The method of the present invention comprehensively considers multi-dimensional quantitative features, not only the sitting posture features and standing posture features of the office scene, but also the eye features, and quantifies the features in a hierarchical manner; through comprehensive analysis of the features, the user's healthy office status can be judged more accurately, thereby improving the accuracy and reliability of health monitoring.

[0052] (2) The method of the present invention helps users perceive their office status and provides health promotion plans. That is, the present invention can identify the user's poor health status in the office scene and remind them, and can also improve the user's health status through the linkage adjustment of the desk and chair.

[0053] (3) This technical solution has broad application prospects. It can be used not only in offices but also in other places such as homes, and can be combined with other health monitoring technologies, such as heart rate monitoring and blood pressure monitoring, to achieve more comprehensive health management. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 is a flow chart of the method of the present invention;

[0055] Figure 2 This is a flow chart of the office status identification and judgment method of the present invention;

[0056] Figure 3 The present invention provides a method for status reminder and health promotion in a sitting office scene;

[0057] Figure 4 This is a status reminder and health promotion method in a standing office scene of the present invention. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0059] like Figure 1 As shown, a method for identifying office status and promoting healthy office work includes the following steps:

[0060] Step 1: Information collection and feature point extraction. This includes the following:

[0061] Use scene information collection devices to collect user office scene information: use office chair pressure sensor devices to collect office sitting posture information, use head and shoulder position collection devices (such as cameras) to collect office standing posture information, and use eye feature collection devices (such as simplified fundus cameras) to collect eye information; and extract feature points of user office scene, sitting posture, standing posture, and eyes. Among them, the feature points of office scene include facial orientation, gaze direction, and screen gaze duration; the sitting posture feature points include the pressure value collected by each grid of the grid pressure sensor; the standing posture feature points include the distance between the body and the edge of the table, the distance between the face and the vertical line of the edge of the table, the distance between the waist and back and the vertical line of the edge of the table, and the line connecting the shoulders and the perpendicular line of the pupil; the eye feature points include the pupil, eyelid, conjunctiva, and gaze.

[0062] Step 2: Determine the office scene.

[0063] The preset office scene judgment rule is used to judge the user's office scene based on the scene information, sitting posture information, standing posture information and eye information. The office scene judgment rule is: if the angle between the facial orientation and the gaze direction is within the screen gaze threshold range [0, 45°], it is judged as staring at the screen. Under this premise, if the screen gaze duration reaches the threshold, the scene is judged to be an office scene; if the screen gaze duration does not reach the threshold, the scene is judged to be a non-office scene.

[0064] The angle between the facial orientation and the gaze direction can be obtained through the existing intelligent system, and the collected images and features are processed and analyzed; that is, the camera collects the facial orientation direction and the gaze direction, and an angle is generated between the two directions.

[0065] Step 3: Analysis of sitting posture, standing posture and eye features.

[0066] (1) Calculate scene information based on the user's office scene feature points; the scene information includes the user's facial orientation, gaze direction, and screen gaze duration.

[0067] The user office scene is obtained through the following methods:

[0068] Obtain facial feature points in office scenes;

[0069] Extract facial orientation, gaze direction, screen gaze start time, and screen gaze end time;

[0070] According to the facial orientation and the gaze direction, the angle between the facial orientation and the gaze direction is calculated to determine whether the user is looking at the screen;

[0071] The screen gaze duration is calculated based on the start time and the end time of the screen gaze.

[0072] The information obtained in the above steps can be used as a basis for judging, monitoring, and adjusting whether the office is healthy in subsequent steps.

[0073] (2) Calculating sitting posture information based on the user's office sitting posture feature points; the sitting posture information includes leaning forward, hunching over, half-lying down, and crossing legs.

[0074] Sitting posture information can be obtained by:

[0075] Get sitting posture feature points;

[0076] Extract the pressure data of each coordinate point based on the sitting posture feature points;

[0077] Based on the pressure data of the coordinate points, it is judged whether the sitting posture is leaning forward, hunched over, half lying down, or crossing legs, and the degree of each sitting posture feature is marked as 0-5, where 0-5 is the degree value, 0 means that the feature does not appear, and 1-5 increases to indicate that it is getting more and more serious.

[0078] The above information can be used as a basis for judging, monitoring and adjusting whether office work is healthy in subsequent steps.

[0079] (3) Calculate standing posture information based on the user's office standing posture feature points; the standing posture information includes leaning forward, bending over, and hunching over.

[0080] The posture information can be obtained by the following calculation method:

[0081] Get standing posture feature points;

[0082] Extract the coordinate points of shoulders, pupils and other parts based on the standing posture feature points;

[0083] Based on the coordinate points, the standing posture is judged as to whether the body is leaning forward, hunched over, and the degree of each standing posture feature, which is marked as 0-5, where 0-5 is the degree value, 0 means that the feature does not appear, and 1-5 increases as it becomes more and more serious.

[0084] The above information can be used as a basis for judging, monitoring and adjusting whether office work is healthy in subsequent steps.

[0085] (4) Calculating eye information based on the user's eye feature points; the eye information includes the distance from the eye to the screen, blinking frequency, conjunctival congestion, and dull eyesight.

[0086] Eye information can be obtained through the following methods:

[0087] Obtain eye image feature points;

[0088] The distance from the eye to the screen, blinking frequency and the quantitative level of these eye features are calculated based on the feature points of the eye image, and are marked as 0-5, where 0 is the degree value, 0 means that the feature does not appear, and 1-5 increases to indicate that it is getting more and more serious.

[0089] Based on the feature points of the eye image, it is judged whether the eye has conjunctival congestion, dull eyesight and the degree of the feature, which is marked as 0-5, where 0-5 is the degree value, 0 means that the feature does not appear, and 1-5 increases to indicate that it is getting more and more serious.

[0090] The above information can be used as a basis for judging, monitoring and adjusting whether office work is healthy in subsequent steps.

[0091] Step 4: Identify office health status.

[0092] The preset office health status judgment rule is used to judge the user's office health status according to the office scene information, sitting posture information, standing posture information and eye information.

[0093] The rules for judging office health status are:

[0094] (1) When the scene is an office scene:

[0095] like Figure 2 As shown in the figure, the preset healthy office status is divided into: very healthy, healthy, general, unhealthy, and very unhealthy, and corresponds to the corresponding threshold range. According to the calculated total score of the sitting posture information, standing posture information, and eye information, the healthy office status can be judged into five states: very healthy, healthy, general, unhealthy, and very unhealthy.

[0096] When the total feature score is judged as very healthy, healthy or general eye health status, if 1 or 2 indicator feature levels reach the maximum threshold, it is judged as an unhealthy working state; if 3 or more feature levels reach the maximum threshold, it is judged as a very unhealthy working state. ， The total score of all the user's characteristics is 16 points, which is generally healthy. However, if the score for leaning forward is 5, then it is judged as unhealthy; or if the score for leaning forward, crossing legs, and hunching over are all 5, then it is judged as very unhealthy.

[0097] When two eye features reach the maximum threshold, it is considered an unhealthy working state; when three or more eye features reach the maximum threshold, it is considered a very unhealthy working state. For example, the total score of all features of the user is 18 points, which is generally healthy, but the score of blinking frequency and eye congestion is 5, which is considered unhealthy.

[0098] (2) When the scenario is a non-office scenario:

[0099] Under this condition, it is impossible to determine the healthy working state. The angle between the facial orientation and the gaze direction must be within the threshold range of the screen gaze, and the acquisition and monitoring time of the eye and sitting / standing posture features must be further extended. When the screen gaze time reaches the threshold, the healthy working state can be determined by the eye and sitting / standing posture features.

[0100] In summary, through the above-mentioned feature extraction and comprehensive analysis of office scene information, sitting posture information, standing posture information and eye information, the healthy office status can be effectively identified and continuously monitored.

[0101] Step 5: Reminder and improvement of adverse health conditions.

[0102] Using the preset office health status adjustment rules, based on the office scene information, sitting posture information, standing posture information and eye information, the user is reminded to make adjustments, or the height / working mode of the office desk and chair is actively adjusted to improve the user's bad office habits and allow the user to have a healthy office state.

[0103] The adjustment rules for office health status are:

[0104] When the scene is an office scene, under this condition, it is further determined whether the person is standing or sitting.

[0105] (1) Figure 3 As shown, when the scene is a standing office state, it is detected whether the desktop height is suitable for the user's height. If it is suitable, the height is not adjusted; if it is not suitable, the user is reminded that the desktop height is not suitable, and the lifting desktop is actively controlled to adjust to the appropriate height.

[0106] Furthermore, it determines whether the user has been standing for a long time. If not, no reminder is given, no height adjustment is made, and the user's standing working status continues to be monitored; if yes, the user is reminded that the user has been standing for a long time, and the desktop height is actively adjusted to a sitting working height suitable for the user's height.

[0107] Furthermore, the chair is adjusted to a working mode, the chair is controlled to be adjusted to a height suitable for the height of the user, the chair back is controlled to return to the normal position, and the lumbar support is controlled to actively support the lumbar.

[0108] Furthermore, it determines whether the standing posture is healthy and reasonable. It detects whether there are bad standing postures such as leaning forward, hunching over, and hunching over, and their characteristic values. If a bad standing posture occurs and the characteristic value exceeds the preset reminder threshold, the user is reminded to pay attention and is advised to make adjustments.

[0109] Furthermore, the eye characteristics are detected and judged. When the detection finds bad eye characteristics, the eye information collection device flashes to remind the user and suggest the user to make adjustments. If it is detected that the distance between the eyes and the screen exceeds the warning threshold, the light flashes to remind the user and control the chair to move forward and backward to adjust to the appropriate eye distance; if it is detected that the blinking frequency exceeds the threshold, or the congestion characteristic value exceeds the threshold, or the dull eyes exceed the threshold, the eye information collection device flashes to attract the user's attention and suggest the user to take a short break to relieve the bad eye symptoms.

[0110] (2) Figure 4 As shown, when the scene is a sitting office state, it is detected whether the height of the table and chair is suitable for the user's height. If it is suitable, the height is not adjusted; if it is not suitable, the user is reminded that the height of the table and chair is not suitable, and the table and chair are actively controlled to be adjusted to a suitable height.

[0111] Furthermore, it determines whether the user has been sitting for a long time. If so, the user is reminded that the user has been sitting for a long time, and the desktop can be adjusted to a height suitable for the user's height to allow the user to stand and work; the chair can also be adjusted to a resting state: lower the height of the chair back and retract the lumbar support; the chair can also be adjusted to a massage mode: the lumbar support massages. If not, no reminder is given, the height is not adjusted, and the user's sitting working state continues to be monitored.

[0112] Furthermore, it determines whether the sitting posture is healthy and reasonable. It detects whether there are bad sitting postures such as leaning forward, hunching over, half-lying, and crossing legs, and their characteristic values. If a bad sitting posture occurs and the characteristic value exceeds the preset reminder threshold, the user is reminded and advised to make adjustments.

[0113] Furthermore, the eye characteristics are detected and judged. When the detection finds bad eye characteristics, the eye information collection device flashes to remind the user and suggest the user to make adjustments. If the detection finds that the distance between the eye and the screen exceeds the warning threshold, or the blinking frequency exceeds the threshold, or the congestion characteristic value exceeds the threshold, or the dull eyes exceed the threshold, the eye information collection device flashes to attract the user's attention and suggest the user to take a short break to relieve the bad eye symptoms.

[0114] Step 6: Dynamic monitoring and identification analysis.

[0115] Use the scene information collection device to continuously collect user office scene information. If it is judged to be an office scene, the office chair pressure sensor device will continue to collect office sitting posture information, or the head and shoulder position collection device will be used to collect office standing posture information, and the eye feature collection device will be used to collect eye information, extract feature points for further feature recognition and judgment, so as to continuously monitor and identify the user's health status in the office scene, provide health status reminders, and provide targeted health promotion plans. If it is judged to be a non-office scene, the office status will not be continuously monitored and analyzed.

[0116] According to the above method, the present invention continuously identifies the level of the user's working status and quantifies the severity while the user is working, and also provides health promotion suggestions and adjustment measures for poor working status, which can help users improve their healthy working level and improve their office efficiency.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying office status and promoting healthy office, characterized in that: The following steps are involved: S1. Information collection and feature point extraction; S2. Office scene judgment; S3. Analysis of sitting posture, standing posture and eye features; S4. Office health identification status; S5. Reminder and improvement of adverse health conditions; S6 dynamic monitoring and identification analysis.

2. A method for identifying office status and promoting healthy office according to claim 1, characterized in that: The step S1 includes the following contents: Collecting user office scene information using a scene information collection device: collecting office sitting posture information using an office chair pressure sensor device, collecting office standing posture information using a head and shoulder position collection device, and collecting eye information using an eye feature collection device; Extract feature points of the user's office scene, sitting posture, standing posture, and eyes.

3. A method for identifying office status and promoting healthy office according to claim 2, characterized in that: In step S2, the preset office scene judgment rule is used to judge the user's office scene based on the scene information, sitting posture information, standing posture information and eye information; wherein, the office scene judgment rule is: if the angle between the facial orientation and the gaze direction is within the threshold range of staring at the screen, it is judged as staring at the screen. Under this premise, if the screen staring time reaches the threshold, the scene is judged to be an office scene; if the screen staring time does not reach the threshold, the scene is judged to be a non-office scene.

4. The method for identifying office status and promoting healthy office work according to claim 2, characterized in that: The step S3 includes the following contents: Calculate scene information based on user office scene feature points; The scene information includes the user's facial orientation, gaze direction, and duration of looking at the screen; Calculate the sitting posture information according to the user's office sitting posture feature points; the sitting posture information includes leaning forward, hunching over, half lying down, and crossing legs; Calculate the standing posture information according to the user's office standing posture feature points; the standing posture information includes leaning forward, bending over and hunching over; Eye information is calculated based on the user's eye feature points; the eye information includes the distance from the eye to the screen, blinking frequency, conjunctival congestion, and dull eyes.

5. The method for identifying office status and promoting healthy office according to claim 2, characterized in that: In the step S4, the user's office health status is determined using a preset office health status determination rule according to the office scene information, sitting posture information, standing posture information and eye information; The office health status judgment rule is as follows: according to the preset healthy office status, it is divided into very healthy, healthy, general, unhealthy, and very unhealthy, and corresponds to the corresponding threshold range. According to the calculated total score of the sitting posture information, standing posture information, and eye information, the healthy office status can be judged into five states: very healthy, healthy, general, unhealthy, and very unhealthy. When the total feature score is judged as very healthy, healthy or general eye health status, if 1 or 2 feature levels reach the maximum threshold, it is judged as an unhealthy working state; if 3 or more feature levels reach the maximum threshold, it is judged as a very unhealthy working state; When two eye feature levels reach the maximum threshold, it is judged as an unhealthy working state; when three or more eye feature levels reach the maximum threshold, it is judged as a very unhealthy working state.

6. A method for identifying office status and promoting healthy office work according to claim 2, characterized in that: In step S5, the preset office health status adjustment rules are used to remind the user to make adjustments based on the office scene information, sitting posture information, standing posture information and eye information, or actively adjust the height / working mode of the office desk and chair to improve the user's bad office habits and allow the user to have a healthy office state.

7. The method for identifying office status and promoting healthy office according to claim 2, characterized in that: The user office scene is obtained in the following way: Obtain facial feature points in office scenes; Extract facial orientation, gaze direction, screen gaze start time, and screen gaze end time; According to the facial orientation and gaze direction, the angle between the facial orientation and gaze direction is calculated to determine whether the user is looking at the screen; The screen gaze duration is calculated based on the start time and the end time of the screen gaze.

8. The method for identifying office status and promoting healthy office work according to claim 2, characterized in that: The sitting posture information is obtained by: Get sitting posture feature points; Extract the pressure data of each coordinate point based on the sitting posture feature points; Based on the pressure data of the coordinate points, determine whether the sitting posture is leaning forward, hunching over, half-lying, crossing legs, and the degree of each sitting posture feature, marked as 0-5.

9. The method for identifying office status and promoting healthy office according to claim 2, characterized in that: The standing posture information is obtained by the following calculation method: Get standing posture feature points; Extract the coordinate points of shoulders, pupils and other parts based on the standing posture feature points; Based on the coordinate points, determine whether the standing posture is leaning forward, hunched over, and the degree of each standing posture feature, marked as 0-5.

10. The method for identifying office status and promoting healthy office work according to claim 2, characterized in that: The eye information is obtained by: Obtain eye image feature points; Based on the feature points of the eye image, the distance from the eye to the screen, the blinking frequency, and the quantization level of these eye features are calculated, marked as 0-5; Based on the feature points of the eye image, determine whether the eye has conjunctival congestion, dull eyesight, and the degree of the feature, marked as 0-5.

Images