An eye-detection method, system, and readable storage medium

Abstract

This invention provides an eye use detection method, system, and readable storage medium. The eye use detection method includes: activating a camera module and detecting key points of the human body using a MediaPipe algorithm to obtain first information; calculating coordinates based on the first information to obtain a calculation result; acquiring depth information from the first information through the camera module and determining whether a user exists based on the calculation result and depth information; if a user exists, obtaining the user's facial pose information based on the calculation result, determining the user's usage scenario based on the facial pose information, and determining whether the user's eyes are detected based on the depth information; if the user's eyes are detected, determining whether the user is using their eyes incorrectly based on the usage scenario and pose information; if incorrect eye use is determined, alerting the user; if the user's eyes are not detected, using a support vector machine classification method for judgment; if the user's eyes are not detected, deactivating the camera module and issuing an alert. This invention improves the intelligence level of eye use reminder devices.

Description

Technical Field

[0001] This invention relates to the field of eye use detection technology, and more specifically, to an eye use detection method, system, and readable storage medium. Background Technology

[0002] With the improvement of living standards and science and technology, many technologies are widely used in daily life. In daily study and life, students' eye habits have always been particularly important. Due to problems with eye habits, myopia is very common. At present, there are many devices that remind you to use your eyes correctly. However, the problems with existing devices are that the detection is not intelligent enough, the scenarios are fixed, so the reminders are limited or excessive in actual use, and they are not convenient to use. Summary of the Invention

[0003] Therefore, embodiments of the present invention provide an eye use detection method, system, and readable storage medium, which improves the intelligence level of eye use reminder devices.

[0004] To address the aforementioned problems, this invention provides an eye-use detection method, comprising: activating a camera module and detecting key points of the human body using a MediaPipe algorithm to obtain first information; performing coordinate calculations based on the first information to obtain calculation results; acquiring depth information from the first information using the camera module, and determining whether a user exists based on the calculation results and depth information; if a user exists, obtaining the user's facial pose information based on the calculation results, determining the user's usage scenario based on the facial pose information, and determining whether the user's eyes are detected based on the depth information; if the user's eyes are detected, determining whether the user is using their eyes incorrectly based on the usage scenario and pose information, and alerting the user if the user is using their eyes incorrectly; if the user's eyes are not detected, using a support vector machine classification method for judgment; if the user's eyes are not detected, deactivating the camera module and issuing an alert.

[0005] Compared to existing technologies, the technical effects achieved by this solution are as follows: By using the MediaPipe algorithm to detect key points on the human body and calculating their positions using coordinates, the key points can be located and acquired more accurately. Simultaneously, by using depth information and coordinate calculation results to obtain facial pose information, the acquisition of facial pose information becomes more convenient, faster, and more accurate. This facilitates more precise judgment of the user's subsequent usage scenarios, leading to better assessment of whether the user's current eye use is correct. Furthermore, by incorporating eye detection, the system enhances scenario assessment, preventing the inability to detect the user's eyes and thus avoiding misjudgments. Overall, the eye use detection is more intelligent and comprehensive, better helping users correct incorrect eye habits and protecting their vision.

[0006] In one embodiment of the present invention, the first information obtained by detecting human body key points using the mediapipe algorithm includes: human body key points include the eyes, nose and shoulders of the human body.

[0007] Compared with existing technologies, the technical effects achieved by this solution are as follows: By setting key points for acquiring the eyes, nose, and shoulders of the human body, it is possible to detect eye use by combining shoulder key points with the eye use detection. The shoulder key points help users locate the eye position and head posture, making the overall eye use detection more reasonable and intelligent, and more accurate, thus better ensuring that users maintain correct eye use habits.

[0008] In one embodiment of the present invention, the coordinate calculation based on the first information and the calculation result further includes: the mediapipe algorithm directly generating the corresponding coordinates based on the eyes, nose and shoulders; and calculating multiple Euler angles corresponding to the eyes, nose and shoulders based on the coordinates.

[0009] Compared with existing technologies, the technical effects achieved by adopting this technical solution are as follows: By setting up the Mediapipe algorithm to directly generate corresponding coordinates for the eyes, nose, and shoulders, and calculating Euler angles accordingly, it is easier to locate the user's eyes later, thereby better obtaining the user's current posture, and thus better reminding and understanding the user's eye habits, ensuring that the user uses their eyes correctly and protecting their vision from damage.

[0010] In one embodiment of the present invention, the user's facial pose information is obtained based on the calculation results. The user's usage scenario is determined based on the facial pose information. The user's eye level angle is set to 0 degrees. When the Euler angle of the eye tilt direction is in the range of 15-45 degrees, the current user usage scenario is defined as writing with the head down. When the Euler angle of the eye tilt direction is in the range of 0-30 degrees, the current user usage scenario is reading with the head up.

[0011] Compared with existing technologies, the technical effects achieved by adopting this technical solution are as follows: by classifying and defining the Euler angle range of the user's eye pitch direction, the user's eye use in two different scenarios can be distinguished, and different reminder restrictions can be given to the user in different usage scenarios. This makes the entire detection system more reasonable and intelligent, and ensures that the user can maintain good eye use habits in different scenarios.

[0012] In one embodiment of the present invention, determining whether a user is using their eyes incorrectly based on the usage scenario and posture information, and if it is determined that the user is using their eyes incorrectly, the reminder to the user further includes: defining the Euler angle range of the eye pitch direction when the user is writing with their head down as a first range, and the Euler angle range of the eye pitch direction when the user is reading with their head up as a second range; when it is determined that the user is using their eyes incorrectly when the user is writing with their head down, and the Euler angle range of the eye pitch direction is less than the second range for a first time, the user is determined to be using their eyes incorrectly; when it is determined that the user is reading with their head up, and the Euler angle range of the eye pitch direction is less than the first range for a first time, the user is determined to be using their eyes incorrectly.

[0013] Compared with existing technologies, the technical effects achieved by this solution are as follows: When writing with the head down, if the Euler angle range in both eyes in the tilt direction is less than the second range in the first short time, it indicates that the user is using their eyes incorrectly. This means that the current head tilt angle is too small, the eye use is incorrect, and it has a significant impact on vision, easily causing vision problems. At the same time, when the user scenario is determined to be reading with the head up, if the Euler angle range in both eyes in the tilt direction is less than the first range in the first short time, it indicates that the user is using their eyes incorrectly. This means that the current reading angle is too large, it is not suitable for reading, and it is easy to cause myopia. Therefore, by using two ranges, the current user scenario can be quickly determined, and eye use judgment can be made simultaneously, making the whole system more intelligent.

[0014] In one embodiment of the present invention, if the user's eyes are not detected, the judgment using the support vector machine classification method further includes: pre-inducing the user to continuously perform the action of looking down at a book, and continuously reducing the downward angle of the head until the camera module can no longer capture the user's eyes; the mediapipe algorithm collects multiple sets of key points of the user's body in the above process to obtain multiple sets of second information; classifying the multiple sets of second information into two categories, incorrect eye use and correct eye use, through the support vector machine; and judging whether the current user is using their eyes correctly based on the multiple sets of second information.

[0015] Compared with existing technologies, the technical effects achieved by this solution are as follows: By pre-acquiring and storing multiple sets of different head postures of users, the system can use this data to determine the current user's head posture and eye use when the user's eye condition cannot be directly obtained. This avoids misjudgments caused by the inability to obtain eye condition data, which could lead to the inability to correct the user's incorrect eye use in time and thus affect the user's vision. In this way, the intelligence level of the entire detection method is improved, making it easier to better protect the user's vision.

[0016] In one embodiment of the present invention, determining whether the current user is using their eyes correctly based on multiple sets of second information further includes: when the user is actually using the device, comparing the first information with multiple sets of second information to determine the classification of the current first information and obtain a classification result; determining whether the current user is using their eyes correctly based on the classification result; and if the user is determined to be using their eyes incorrectly, providing a reminder to use their eyes.

[0017] Compared with existing technologies, the technical effects achieved by this solution are as follows: by comparing the user's actual eye usage information with pre-collected second information, the user's current eye usage status can be determined quickly even when the user's eye condition cannot be obtained, thus protecting the user's vision. This makes the detection process more comprehensive and intelligent, and better protects the user's eyesight.

[0018] In one embodiment of the present invention, an eye use detection system is also provided. The eye use detection method of any of the above is applied to the eye use detection system. The eye use detection system includes: an identification module for identifying the user and key points of the human body; a calculation module for calculating coordinates and surface pose information; and a judgment module for judging whether the user is using their eyes correctly.

[0019] This eye use detection system has all the features of the aforementioned eye use detection methods, and therefore has the same technical effect, which will not be elaborated here.

[0020] The present invention also provides a readable storage medium on which computer-executable instructions are stored, wherein, when the computer-executable instructions are run by a processor, the device on which the storage medium is located controls the execution of steps implementing the eye detection method as described above.

[0021] By adopting the technical solution of the present invention, the following technical effects can be achieved:

[0022] (1) By setting the MediaPipe algorithm to detect key points of the human body and calculating the position of key points using coordinates, the position of key points of the human body can be more accurately located and obtained. At the same time, by using depth information and coordinate calculation results to obtain face pose information, face pose information acquisition is more convenient, faster and more accurate, which facilitates more accurate judgment of the user's subsequent usage scenario, and thus better judges whether the user's current eye use is correct. In addition, by setting the detection of the user's eyes, the judgment scenario is increased, avoiding the inability to judge the user's eye use habits due to the failure to detect the user's eyes, thus causing misjudgment. This makes the overall user eye use detection more intelligent and comprehensive, which can better help users correct incorrect eye use habits and protect the user's vision.

[0023] (2) By setting the Euler angle range of both eyes in the tilt direction to be smaller than the second range in the first time when the user is writing with their head down, it is judged that the user is using their eyes incorrectly. This indicates that the current head-down angle is too small and the user is using their eyes incorrectly, which has a significant impact on vision and is likely to cause vision problems. At the same time, by setting the Euler angle range of both eyes in the tilt direction to be smaller than the first range in the first time when the user is judging the usage scenario as reading with their head up, it is judged that the user is using their eyes incorrectly. This indicates that the current reading angle is too large and is not suitable for reading, which is likely to cause myopia. Therefore, by using the two ranges, the current user's usage scenario can be quickly judged and the user's eye use judgment can be made, making the whole system more intelligent.

[0024] (3) By pre-acquiring and storing different head postures of multiple users, the user's head posture and eye use can be judged by these data when the user's eye condition cannot be directly obtained. This avoids misjudgment when the eye condition cannot be obtained, which would prevent timely correction of the user's incorrect eye use and thus affect the user's vision. This improves the intelligence of the entire detection method and makes it easier to better protect the user's vision. Attached Figure Description

[0025] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings to be used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a flowchart illustrating an eye use detection method provided in Embodiment 1 of the present invention.

[0027] Figure 2 This is a schematic diagram of an eye use detection system provided in Embodiment 1 of the present invention.

[0028] Figure 3 This is a schematic diagram of the structure of a readable storage medium according to Embodiment 2 of the present invention.

[0029] Explanation of reference numerals in the attached figures:

[0030] 100 is the eye detection method; 110 is the recognition module; 120 is the calculation module; 130 is the judgment module; 200 is the readable storage medium; 210 is the computer-executable instruction. Detailed Implementation

[0031] To make the above-mentioned objectives, features, and advantages of the present invention more apparent and understandable, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] [First Embodiment]

[0033] See Figures 1-2 The present invention provides a method for detecting eye use, comprising:

[0034] Step S100: Turn on the camera module and detect key points of the human body using the MediaPipe algorithm to obtain the first information; calculate the coordinates based on the first information to obtain the calculation result; obtain the depth information of the first information through the camera module, and determine whether a user exists based on the calculation result and the depth information;

[0035] Step S200: If it exists, obtain the user's face pose information based on the calculation result, determine the user's usage scenario based on the face pose information, and determine whether the user's eyes are detected based on the depth information.

[0036] Step S300: If the user's eyes are detected, determine whether the user is using their eyes incorrectly based on the usage scenario and pose information. If the user is determined to be using their eyes incorrectly, remind the user. If the user's eyes are not detected, use the support vector machine classification method to make a judgment. If the user's eyes are not detected, turn off the camera module and issue a reminder.

[0037] Specifically, the Mediapipe algorithm detects key points on the human body (the points used for eye detection are the eyes, nose, and shoulders) and calculates their coordinates. Mediapipe directly generates the coordinates of the corresponding points after key point detection. The coordinates of the eyes are calculated separately when calculating Euler angles (calculating Euler angles requires the coordinates of three points), and the average value is calculated during SVM (Support Vector Machine) classification. The average value of the shoulder coordinates is also calculated.

[0038] Furthermore, by using three key facial points (eyes, nose), the current facial pose information (three Euler angles of the head) is estimated. Analyzing this facial pose information allows for the estimation of the user's current usage scenario (whether they are looking down to write or looking up to read). A binocular camera, i.e., a camera module, can acquire depth information in front of the camera. Combining the above information, this patent can analyze the following scenarios: 1. Whether a user is currently using the function; if no user is present, the function automatically closes or prompts the user; if a user is present, the eye detection function is activated. 2. When a user uses the function, the user's usage scenario can be determined, whether they are looking down to write or looking up to read, with different judgment logic applied to different scenarios. 3. Combining the estimated facial pose and depth information, while protecting user privacy (the camera cannot capture desktop information), the user's eye distance can be estimated and a reasonable reminder can be given. The height of the user's current desktop relative to the horizontal eye level is estimated based on the coordinates of the user's shoulder key points, and the distance from the eye to the desktop is estimated based on the Euler angles of the head and the eye coordinates.

[0039] Furthermore, when a user's posture of looking down while writing or reading reaches a certain point, the camera in front may fail to detect the user's eyes correctly. However, in this scenario, the MediaPipe model will still estimate the position of the eyes (based on information from shoulder points to determine the eye position). In this scenario, it cannot correctly estimate the Euler angles of the head or calculate the viewing distance. Therefore, the above methods cannot be used to estimate and remind the user of the viewing distance. Thus, this patent uses an SVM classifier to classify the situation where the user's eyes cannot be detected correctly when looking down into correct and incorrect viewing distances. The steps are as follows: multiple sets of the above key points are collected in advance and classified into two categories, incorrect viewing distance and correct viewing distance, using an SVM. Subsequently, in this scenario, the camera collects the user's above key point information in real time, and the classifier is called to classify the current posture in real time. Combined with judgment logic and timing logic, the system reminds the user whether they should pay attention to the viewing distance.

[0040] Preferably, the system uses the MediaPipe algorithm to detect key points on the human body and calculates their positions using coordinates. This allows for more precise location and acquisition of these key points. Simultaneously, depth information and coordinate calculations are used to obtain facial pose information, making this acquisition more convenient, faster, and more accurate. This facilitates more accurate judgment of the user's subsequent usage scenarios and better assessment of whether the user's eye use is correct. Furthermore, by detecting the user's eyes, the system enhances scenario assessment, preventing the inability to detect the user's eyes and thus avoiding misjudgments. This makes the overall eye use detection more intelligent and comprehensive, better helping users correct incorrect eye habits and protecting their vision.

[0041] Specifically, the Mediapipe algorithm detects key points on the human body, and the first information obtained includes: key points on the human body include the eyes, nose, and shoulders.

[0042] Preferably, by setting key points for acquiring the eyes, nose, and shoulders of the human body, it is possible to detect eye use by combining shoulder key points with the eye use detection. The shoulder key points help users locate the eye position and head posture, making the overall eye use detection of users more reasonable and intelligent, and more accurate, so as to better ensure that users maintain correct eye use habits.

[0043] Specifically, the coordinate calculation based on the first information also includes: the mediapipe algorithm directly generating the corresponding coordinates based on the eyes, nose, and shoulders; and calculating multiple Euler angles corresponding to the eyes, nose, and shoulders based on the coordinates.

[0044] Preferably, by setting up the Mediapipe algorithm to directly generate corresponding coordinates for the eyes, nose, and shoulders, and calculating Euler angles accordingly, it becomes easier to locate the user's eyes later, thereby better obtaining the user's current posture, and thus better reminding and understanding the user's eye habits, ensuring that the user uses their eyes correctly and that their vision is not damaged.

[0045] Specifically, the user's facial pose information is obtained based on the calculation results. The user's usage scenario is determined based on the facial pose information. The user's eye level angle is set to 0 degrees. When the Euler angle of the eye level is between 15 and 45 degrees, the current user usage scenario is defined as writing with the head down. When the Euler angle of the eye level is between 0 and 30 degrees, the current user usage scenario is reading with the head up.

[0046] Preferably, by classifying and defining the Euler angle range of the user's eye tilt direction, the system differentiates the user's eye use in two different scenarios, providing different reminders and restrictions for users in different usage scenarios. This makes the entire detection system more reasonable and intelligent, ensuring that users can maintain good eye habits in different scenarios.

[0047] Specifically, based on the usage scenario and posture information, it is determined whether the user is using their eyes incorrectly. If the user is determined to be using their eyes incorrectly, the reminder to the user also includes: defining the Euler angle range of the eyes in the tilt direction as the first range when the user is writing with their head down, and the Euler angle range of the eyes in the tilt direction as the second range when the user is reading with their head up; when the user is writing with their head down, if the Euler angle range of the eyes in the tilt direction is less than the second range in the first time, the user is determined to be using their eyes incorrectly; when the user is reading with their head up, if the Euler angle range of the eyes in the tilt direction is less than the first range in the first time, the user is determined to be using their eyes incorrectly.

[0048] Preferably, by setting the system so that when writing with the head down, if the Euler angle range of both eyes in the tilt direction is less than the second range in the first short time, it is determined that the user is using their eyes incorrectly. This indicates that the current head tilt angle is too small, the eye use is incorrect, and it has a significant impact on vision, which can easily cause vision problems. At the same time, by setting the system so that when the user scenario is reading with the head up, if the Euler angle range of both eyes in the tilt direction is less than the first range in the first short time, it is determined that the user is using their eyes incorrectly. This indicates that the current reading angle is too large, which is not suitable for reading and can easily cause myopia. Therefore, by using the two ranges, the current user scenario can be quickly determined, and an eye use judgment can be made, making the whole system more intelligent.

[0049] Specifically, if the user's eyes are not detected, the support vector machine classification method is used to determine whether the user's eyes are being captured. This includes: first, having the user continuously look down at a book and gradually lowering the angle of their head until the camera module can no longer capture the user's eyes; the mediapipe algorithm collects multiple sets of key points of the user's body during the above process to obtain multiple sets of secondary information; classifying the multiple sets of secondary information into two categories, incorrect eye use and correct eye use, using the support vector machine; and then determining whether the current user is using their eyes correctly based on the multiple sets of secondary information.

[0050] Preferably, by pre-acquiring and storing multiple sets of different head postures of users, this data can be used to determine the current user's head posture and eye use when the user's eye condition cannot be directly obtained. This avoids misjudgment caused by the inability to obtain eye condition data, which would prevent timely correction of the user's incorrect eye use and thus affect the user's vision. This improves the intelligence level of the entire detection method and facilitates better protection of the user's vision.

[0051] Specifically, determining whether a user is using their eyes correctly based on multiple sets of secondary information also includes: comparing the primary information with multiple sets of secondary information during actual use to determine the classification of the primary information and obtain a classification result; determining whether the user is using their eyes correctly based on the classification result; and reminding the user to use their eyes correctly if incorrect eye use is detected.

[0052] Preferably, the current user's eye condition is determined by comparing the first information from the user's actual use with the second information collected in advance. This allows for a quick assessment of the user's eye condition even when the user's eye condition cannot be obtained, thus protecting the user's vision. This makes the detection process more comprehensive and intelligent, and better protects the user's eyesight.

[0053] Specifically, an eye use detection system 100 is also provided. Any of the eye use detection methods described above is applied to the eye use detection system 100. The eye use detection system 100 includes: an identification module 110 for identifying the user and key points of the human body; a calculation module 120 for calculating coordinates and surface pose information; and a judgment module 130 for judging whether the user is using their eyes correctly.

[0054] In one specific embodiment, the identification module 110, the calculation module 120, and the judgment module 130 work together to implement the eye use detection system method described above, which will not be repeated here.

[0055] The eye use detection system 100 has all the features of the aforementioned eye use detection method, and therefore has the same technical effect, which will not be elaborated here.

[0056] [Second Embodiment]

[0057] See Figure 3 The present invention also provides a readable storage medium 200, on which computer-executable instructions 210 are stored, wherein when the computer-executable instructions 210 are run by a processor, the device where the storage medium is located controls the execution of steps implementing the eye detection method as described above.

[0058] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the functional modules in the various embodiments of the present invention can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0059] If the functionality is implemented as a software module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned readable storage medium 200 includes various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting eye use, characterized in that, include: The camera module is activated, and key points of the human body are detected using the MediaPipe algorithm to obtain initial information. Based on the first information, coordinate calculations are performed to obtain the calculation results; The depth information of the first information is obtained through the camera module, and the presence of a user is determined based on the calculation result and the depth information. If present, the user's facial pose information is obtained based on the calculation result, the user's usage scenario is determined based on the facial pose information, and the user's eyes are determined based on the depth information. If the user's eyes are detected, it is determined whether the user is using their eyes incorrectly based on the usage scenario and the posture information. If it is determined that the user is using their eyes incorrectly, the user is reminded. If the user's eyes are not detected, a support vector machine classification method is used to make a judgment. If it does not exist, the camera module will be turned off and a reminder will be issued; The user is instructed to continuously look down at a book, and the angle of the head downwards is gradually lowered until the camera module can no longer capture the user's eyes; The MediaPipe algorithm collects multiple sets of key points of the user's body during the above process to obtain multiple sets of second information. The multiple sets of second information are classified into two categories, incorrect eye use and correct eye use, using a support vector machine. Based on the multiple sets of second information, determine whether the current user is using their eyes correctly; When the user actually uses it, the first information is compared with the multiple sets of second information to determine the current classification of the first information and obtain the classification result; Based on the classification results, determine whether the current user is using their eyes correctly; If the user is judged to be using their eyes incorrectly, the user will be reminded of the correct eye use.

2. The eye use detection method according to claim 1, characterized in that, The first information obtained by detecting key points of the human body using the MediaPipe algorithm includes: The key points of the human body include the eyes, nose, and shoulders.

3. The eye use detection method according to claim 2, characterized in that, The step of calculating coordinates based on the first information to obtain the calculation result further includes: The Mediapipe algorithm directly generates corresponding coordinates based on the eyes, nose, and shoulders. Calculate multiple Euler angles corresponding to the eyes, nose, and shoulders based on the coordinates.

4. The eye use detection method according to claim 3, characterized in that, The step of obtaining the user's facial pose information based on the calculation result, and determining the user's usage scenario based on the facial pose information, further includes: Set the user's eye level to 0 degrees; When the Euler angle of the eye's tilt direction is within the range of 15-45 degrees, the current user scenario is defined as writing with the head down. When the Euler angle of the eye's tilt direction is within the range of 0-30 degrees, the current user scenario is defined as looking up to read.

5. The eye use detection method according to claim 4, characterized in that, The step of determining whether the user is using their eyes incorrectly based on the usage scenario and the posture information, and if it is determined that the user is using their eyes incorrectly, further includes reminding the user: The Euler angle range of the eye pitch direction when the user is writing with their head down is defined as the first range, and the Euler angle range of the eye pitch direction when the user is reading with their head up is defined as the second range. When the usage scenario is determined to be writing with the head down, if the Euler angle range of the eye pitch direction is less than the second range in the first time, it is determined that the user is using their eyes incorrectly. When the usage scenario is determined to be looking up to read, if the Euler angle range of the eye pitch direction is less than the first range in the first time period, it is determined that the user is using their eyes incorrectly.

6. An eye use detection system, characterized in that, The eye use detection method according to any one of claims 1 to 5 is applied to the eye use detection system, wherein the eye use detection system comprises: The recognition module is used to identify the user and key points of the human body; The calculation module is used to calculate the coordinates and the pose information of the surface part; The judgment module is used to determine whether the user is using their eyes correctly.

7. A readable storage medium, characterized in that, The computer-executable instructions stored on the readable storage medium, when executed by a processor, control the device containing the storage medium to perform the steps of implementing the eye detection method as described in any one of claims 1 to 5.

Images