Intelligent vision detection method

Through intelligent vision detection methods, facial recognition and dynamic distance monitoring technology are used to capture the gaze point in real time and evaluate the credibility, solving the accuracy and efficiency of vision detection in the existing technology, and is suitable for low-cooperation in infants and young children. The population has achieved efficient and accurate vision detection results.

CN119091494AInactive Publication Date: 2024-12-06江苏富翰医疗产业发展有限公司
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411076799.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-06
Publication Date
2024-12-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing vision detection technology cannot automatically capture the gaze position of the examined person at different examination distances in real time, and the examination accuracy and credibility of people with incomplete language and limb expression abilities is low, time-consuming, and affected by subjective factors.

Method used

Intelligent vision detection methods are adopted, including face recognition, distance perception, dynamic distance monitoring, viewpoint monitoring and dichotomy level judgment. The position of the gaze point is captured in real time through the dual-screen collaboration module, and combined with dynamic distance monitoring and credibility evaluation, automatically adjust the inspection. Level and results assessment.

Benefits of technology

Real-time identification of gaze points is achieved, the accuracy and efficiency of vision detection is improved, artificial errors are reduced, and it is suitable for people with low coordination. It provides a variety of mode choices and shortens the inspection time.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119091494A_ABST
    Figure CN119091494A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of vision detection, and discloses an intelligent vision detection method, which comprises the following steps: preparation before detection: information input: a detected person collects a face photo before detection and stores the face photo in a database, and inputs basic information including name, age and gender at the same time; and face verification: performing face detection and face recognition on the detected person, comparing the feature information with the face photo collected in the database, and if the information is consistent, starting the next step. According to the method, the fixation points of the examined person under different examination distances can be recognized in real time, vision result conversion is completed, vision screening results are directly exported, multiple modes are provided for a user to select, and the problems of distance constraint, large manual examination errors, long consumed time and the like in the examination process are solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of vision detection, and in particular to an intelligent vision detection method. Background Art

[0002] Vision testing is the process of evaluating the visual function of the eyes and usually includes testing of multiple aspects of the eyes to determine whether vision is normal, whether there are visual defects, and whether glasses, contact lenses or other vision correction are needed.

[0003] Studies have shown that vision screening is an important medical examination for people of all ages. The cognitive and comprehension levels of different people vary greatly. Especially for children aged 0 to 3 months who have incomplete language and body expression abilities, their cooperation in participating in vision screening is extremely low, which makes it difficult for doctors to accurately capture the subject's gaze behavior on the visual target during manual examination, and the examination process is time-consuming.

[0004] The existing technology focuses on the improvement of vision screening display devices and equipment, the optimization of system database modules, and the optimization of semi-automatic inspection processes. It is unable to capture the gaze position of the examinee at different inspection distances in real time and automatically. The examinee still needs to maintain a fixed inspection distance and complete the vision inspection through the guidance of the inspector and the examinee's independent judgment. Then the inspector manually judges whether the examinee is looking correctly, and then obtains the vision inspection result. The current technology is not truly intelligent and is affected by subjective factors. It is especially difficult to complete the inspection for people who do not have complete language expression and body expression ability, and the accuracy and credibility of the inspection results are not high. Summary of the invention

[0005] The purpose of the present invention is to solve the problems existing in the prior art and to propose an intelligent vision detection method.

[0006] In order to achieve the above object, the present invention is applied to the following technical solutions:

[0007] An intelligent vision detection method comprises the following steps:

[0008] Preparation before testing:

[0009] Information entry: Before the examination, the person being examined takes a facial photo and stores it in the database, and at the same time enters basic information including name, age and gender;

[0010] Face verification: Perform face detection and face recognition on the inspected person, and compare the feature information with the face photos collected in the database. If the information is consistent, the next step can be started;

[0011] Distance perception: The person being examined will be measured in real time, and prompted to adjust the examination distance until it is within the medically recommended distance range;

[0012] Vision test:

[0013] Obtaining the gaze point position of the person being inspected;

[0014] Each sight mark is displayed randomly;

[0015] Compare the coincidence between the gaze point position and the random position where the current sight mark appears. If the gaze point position appears within the sight mark area, it is considered that a sight mark has occurred, and record the inspection information when each sight mark occurs.

[0016] Get the visual acuity result.

[0017] Preferably, when the person to be tested is an infant or young child with a low degree of cooperation, the pre-test preparation and vision test process also includes: playing animated images and audio to attract the attention of the person to be tested, while achieving a reward effect and stimulating the person to be tested's interest in the test.

[0018] Preferably, during the distance sensing, the number of people at the inspected location is detected. If the number of people is greater than one, the vision test is only performed on the nearest inspected person, because the examination of young infants and young children requires the accompaniment of their parents. At the same time, it can also prevent multiple people from accidentally entering the camera and causing the examination to be interrupted.

[0019] Preferably, the specific process of obtaining the gaze point position of the person being tested is: capturing the dynamic features of the face and head of the person being tested in real time, completing the line of sight direction estimation, and calculating the gaze point position based on the actual size and resolution of the display device where the sight mark is located and the real-time distance between the person being tested and the display device, so as to achieve accurate projection of the viewpoint on the display device.

[0020] Preferably, the inspection information includes the position of the gaze point, the duration of gaze, and the real-time distance at the corresponding moment. Finally, the total number of gazes and the total duration of gazes within the current sight mark display time are counted and recorded. One sight mark corresponds to a complete record of inspection information.

[0021] Preferably, it also includes grading the vision test results: setting the initial level according to the visual ability of different age groups, and automatically determining the level of the next test by using a dichotomy method;

[0022] If the current level fails, a lower level check will be conducted until a lower level is successfully passed, and this level is considered the vision level;

[0023] If the current level is successfully passed, a higher level check will be carried out until a certain higher level fails, and the previous level of the level is considered to be the vision level.

[0024] Preferably, the vision value results include CPD vision value, Sellen decimal vision value and 5-point vision value.

[0025] Preferably, the vision test further includes: evaluating whether the cumulative duration of the visual behavior of the visual mark within each single display time period of the visual mark indicates that the gaze behavior is credible, and if it is credible, judging that the inspection of the current visual mark level has passed.

[0026] Preferably, after obtaining the visual acuity value result, the result of the visual acuity test is exported and a report is generated. The display content in the report includes a gaze point-visual mark scatter point comparison chart, a visual acuity value statistical chart, and a visual acuity value.

[0027] An intelligent vision detection system is applied to an intelligent vision detection method, comprising:

[0028] Dual-screen collaboration module: The secondary screen is the detection end screen, which displays all the function pages of vision screening, and the main screen is the screen of the inspected end, which displays different levels of visual marks and animations. The camera is located in the middle above the main screen; the secondary screen page has a separate area, which synchronizes the main screen page and the monitoring screen of the inspected person in real time to achieve dual-screen collaboration;

[0029] Information entry module: Before the inspection, the inspected person takes a facial photo and stores it in the database, and at the same time enters basic information such as name, age, and gender;

[0030] Face recognition module: performs face detection and face recognition on the subject, and compares the feature information with the face photos collected in the database. If the information is consistent, the next step can be started;

[0031] Distance sensing module: used to measure the distance between the person being tested and the main screen in real time before and during the vision test;

[0032] Attention-attracting module: This module can be used selectively. For infants and young children with low cooperation during the examination, animated images and audio can be played before and during the examination to attract the attention of the examinee, and at the same time, the effect of reward can be achieved to stimulate the examinee's interest in the examination.

[0033] Pre-examination distance prompt submodule: If the information is successfully matched in the face recognition module, the distance of the person being examined will be measured in real time, and the person will be prompted to adjust the examination distance until it is within the medically recommended distance range, and then the vision examination can begin;

[0034] Dynamic distance monitoring submodule: Real-time distance measurement during the whole process of vision examination, keeping real-time distance measurement, ensuring that only the person closest to the main screen is examined during the examination, because the examination of young infants and young children requires the company of their parents, and it can also prevent the situation where multiple people accidentally enter the camera and cause the examination to be interrupted;

[0035] Viewpoint monitoring module: calculates the gaze point position of the inspected person;

[0036] Viewpoint judgment module: Combined with the information of the dynamic distance monitoring submodule in the distance perception module, for the inspected person closest to the main screen, the viewpoint position calculated by the viewpoint monitoring module is compared with the random position where the current sight mark appears in real time. If the viewpoint position appears in the sight mark area, it is considered that a sight mark has occurred;

[0037] Level judgment module: used to judge whether the current level of visual mark test has passed and whether it can enter a higher level of test;

[0038] Credibility Assessment Submodule: It is used to assess whether the cumulative duration of the visual behavior of the visual mark within each single display period indicates that the gaze behavior is credible. The assessment standard can be set by the examiner on the secondary screen page and combined with clinical experience;

[0039] Promotion judgment submodule: used to judge whether the current visual standard level inspection can be passed based on the results of the credibility assessment submodule;

[0040] Conventional detection unit: Check from the lowest level step by step until the promotion judgment submodule determines that the current level has failed and locates the vision level to the previous level of the current level;

[0041] Intelligent rapid detection unit: according to the visual ability of different age groups, the initial level is set and the next level of inspection is automatically determined by dichotomy;

[0042] Test result conversion module: used to calculate the visual acuity value of the person being tested according to the test record value of the visual acuity level. The visual acuity results include CPD visual acuity value, Sellen decimal visual acuity value and 5-point visual acuity value;

[0043] Report export module: used to export the results of vision screening. The content displayed in the report includes the comparison chart of fixation point and sight mark, statistical chart of vision value, and vision value. The tester can choose to print it.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] The present invention can identify the gaze point of the person being inspected at different inspection distances in real time, complete the vision result conversion, directly derive the vision screening results, and also provide a variety of modes for users to choose from, solving the problems of distance constraints in the inspection process, large errors in manual inspection, and long time consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A schematic diagram of a flow chart of an intelligent vision detection method proposed by the present invention;

[0047] Figure 2 It is a schematic diagram of the scatter plot distribution of viewpoints relative to the sight mark;

[0048] Figure 3 This is a schematic diagram of an intelligent vision detection system proposed by the present invention.

[0049] In the figure: 1. Dual-screen collaboration module; 2. Information entry module; 3. Face recognition module; 4. Distance perception module; 5. Attention attraction module; 6. Pre-inspection distance prompt sub-module; 7. Viewpoint monitoring module; 8. Viewpoint judgment module; 9. Level judgment module; 10. Conventional detection unit; 11. Intelligent rapid detection unit; 12. Detection result conversion module; 13. Report export module; 14. Dynamic distance monitoring sub-module; 15. Credibility assessment sub-module; 16. Promotion judgment sub-module. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0051] like Figure 1 and Figure 2 As shown, an intelligent vision detection method comprises the following steps:

[0052] Preparation before testing:

[0053] Information entry: Before the examination, the person being examined takes a facial photo and stores it in the database, and at the same time enters basic information including name, age and gender;

[0054] Face verification: Face detection and face recognition are performed on the person being tested, and the feature information is compared with the face photos collected in the database. If the information is consistent, the next step can be started; this step can ensure the reliability of the data of the person being tested, because when there are a large number of people, there may be people with the same name, etc., and the doctor's one-to-one verification is inefficient. The use of face recognition to verify information before the test can ensure that the test results match the person being tested and ensure the validity of the results;

[0055] Distance perception: The person being tested will be measured in real time, and prompted to adjust the inspection distance until it is within the medically recommended distance range, thereby ensuring the validity and accuracy of the test data;

[0056] Vision test:

[0057] Obtaining the gaze point position of the person being inspected;

[0058] Each sight mark is displayed randomly;

[0059] Compare the coincidence between the gaze point position and the random position where the current sight mark appears. If the gaze point position appears within the sight mark area, it is considered that a sight mark has occurred, and record the inspection information when each sight mark occurs.

[0060] Get the visual acuity result.

[0061] Compared with the existing technology, this intelligent vision detection method can identify the gaze point of the person being tested at different inspection distances in real time, complete the vision result conversion, directly export the vision screening results, and provide a variety of modes for users to choose from, solving problems such as distance constraints in the inspection process, large errors in manual inspection, and long time consumption.

[0062] In this embodiment, when the person being tested is an infant or young child with low cooperation, the pre-test preparation and vision test process also includes: playing animated images and audio to attract the attention of the person being tested, while achieving a reward effect and stimulating the person's interest in the test.

[0063] In this embodiment, during the distance sensing, the number of people at the inspected location is detected. If the number of people is greater than one, the vision test is only performed on the nearest inspected person. Because the examination of young infants and young children requires the accompaniment of their parents, this can prevent the incorrect identification of the inspected person and also prevent the inspection from being interrupted due to multiple people accidentally entering the camera.

[0064] In this embodiment, the specific process of obtaining the gaze point position of the inspected person is: capturing the dynamic features of the face and head of the inspected person in real time, completing the line of sight direction estimation, and calculating the gaze point position according to the actual size and resolution of the display device where the sight mark is located and the real-time distance between the inspected person and the display device, so as to realize the accurate projection of the viewpoint on the display device;

[0065] The above process first detects and tracks the face and head through the camera, using computer vision technology, such as convolutional neural network (CNN) in deep learning, to capture and track the position, posture and movement of the face and head in real time, and identify facial feature points, such as eyes, nose, mouth, etc., as well as the tilt and rotation angle of the head;

[0066] Then the eyes and pupils are located. This is mainly based on facial tracking. The positions of the eyes and pupils are further located. This process can be achieved by analyzing the image features of the eye area, such as pupil edges, iris color changes, etc.

[0067] Finally, the line of sight direction is estimated. In the absence of medium changes (such as air layers of different densities), the line of sight is straight. Based on the dynamic characteristics of the eyes and head, the line of sight direction is calculated, and according to the parameters of the display device, the point where the line of sight falls on the screen is calculated.

[0068] In this embodiment, the inspection information includes the position of the gaze point, the duration of gaze, and the real-time distance at the corresponding moment. Finally, the total number of gazes and the total duration of gaze within the current sight mark display time are counted and recorded. One sight mark corresponds to a complete record of inspection information.

[0069] In this embodiment, it also includes grading the vision test results: according to the visual ability of different age groups, the initial level is set, and the next level of examination is automatically determined by using a dichotomy method;

[0070] If the current level fails, a lower level check will be conducted until a lower level is successfully passed, and this level is considered the vision level;

[0071] If the current level is successfully passed, a higher level check will be carried out until a certain higher level fails, and the previous level of the level is considered to be the vision level.

[0072] For example, using the Snellen eye chart as the standard, set the visual acuity level from 20 / 200 to 20 / 20, with the midpoint being 20 / 100;

[0073] Ask the subject to try to identify the smallest character at the intermediate vision level;

[0074] If the subject can recognize the characters at the middle point, the range will be adjusted from the middle point to the high end of the range. For example, if the middle point is 20 / 100 and the subject can recognize it, the new range will be 20 / 100 to 20 / 20 (i.e., the level is successfully passed and entered the next level).

[0075] If the subject cannot recognize the characters at the midpoint, the range is adjusted from the low end of the range to the midpoint. For example, if the midpoint is 20 / 100 and the subject cannot recognize it, the new range will be 20 / 200 to 20 / 100 (i.e., the pass fails and enters the next level).

[0076] In this embodiment, the vision value results include CPD vision value, Sellen decimal vision value and 5-point vision value.

[0077] In this embodiment, the vision test also includes: evaluating whether the cumulative duration of the visual behavior of the visual mark within each single display time period indicates that the gaze behavior is credible. If it is credible, it is determined that the inspection of the current visual mark level has passed. Because in actual testing, the person being tested may lie, resulting in unreliable results, and this embodiment is fully electronic, the test results are credible and efficient.

[0078] In this embodiment, after obtaining the vision value result, the result of the vision test is exported and a report is generated. The display content in the report includes a gaze point-visual mark scatter point comparison chart, a vision value statistical chart, and a vision value.

[0079] like Figure 3 As shown, this embodiment also discloses an intelligent vision detection system, which is applied to an intelligent vision detection method, including:

[0080] Dual-screen collaboration module 1: includes a main screen, a secondary screen, a camera, and a built-in processor. The secondary screen is the detection end screen, which displays all the function pages of vision screening. The main screen is the screen of the inspected end, which displays different levels of visual marks and animations. The camera is located in the middle above the main screen. The secondary screen page has a separate area to synchronize the main and secondary screen pages and the monitoring screen of the inspected person in real time to achieve dual-screen collaboration. The processor is used to realize data processing and interaction of all modules of this system.

[0081] Information entry module 2: Before the inspection, the inspected person takes a facial photo and stores it in the database, and at the same time enters basic information such as name, age, and gender;

[0082] Face recognition module 3: Perform face detection and face recognition on the subject, compare the feature information with the face photos collected in the database, and if the information is consistent, proceed to the next step;

[0083] Distance sensing module 4: used to perform real-time distance measurement before the vision test and during the entire vision test, and to measure the distance between the person being tested and the main screen;

[0084] Attention Attraction Module 5: This module can be used selectively. For infants and young children with low cooperation during the examination, animated images and audio can be played before and during the examination to attract the attention of the examinee, and at the same time achieve the effect of reward and stimulate the examinee's interest in the examination;

[0085] Pre-examination distance prompt submodule 6: If the information comparison is successful in the face recognition module 3, the distance of the person being examined will be measured in real time, and the person will be prompted to adjust the examination distance until it is within the medically recommended distance range, and then the vision examination can begin;

[0086] Dynamic distance monitoring submodule 14: Real-time distance measurement during the whole process of vision examination, maintaining real-time distance measurement, ensuring that during the examination, only the person being examined who is closest to the main screen is examined for vision, because the examination of young infants and young children requires the company of their parents, and it can also prevent multiple people from entering the camera by mistake, causing the examination to be interrupted;

[0087] Viewpoint monitoring module 7: calculates the gaze point position of the inspected person;

[0088] Viewpoint judgment module 8: Combined with the information of the dynamic distance monitoring submodule in the distance perception module 4, for the person being inspected who is closest to the main screen, the viewpoint position calculated by the viewpoint monitoring module 7 is compared with the random position where the current sight mark appears in real time. If the viewpoint position appears in the sight mark area, it is considered that a sight mark has occurred;

[0089] Level judgment module 9: used to judge whether the current level of visual mark test has passed and whether it can enter a higher level of test;

[0090] Credibility evaluation submodule 15: used to evaluate whether the cumulative duration of the visual behavior of the visual mark within each single display period indicates that the gaze behavior is credible. The evaluation standard can be set by the examiner on the secondary screen page and combined with clinical experience;

[0091] Promotion judgment submodule 16: used to judge whether the current sight mark level inspection can be passed based on the result of the credibility assessment submodule;

[0092] Conventional detection unit 10: checks from the lowest level step by step until the promotion judgment submodule determines that the current level has failed and locates the vision level to the previous level of the current level;

[0093] Intelligent rapid detection unit 11: Set the initial level according to the visual ability of different age groups, and use the dichotomy method to automatically determine the next level of inspection;

[0094] The test result conversion module 12 is used to calculate the visual acuity value of the person being tested according to the test record value of the visual acuity level, and the visual acuity result includes the CPD visual acuity value, the Sellen decimal visual acuity value and the 5-point visual acuity value;

[0095] Report export module 13: used to export the results of vision screening. The display content in the report includes a comparison chart of fixation points and sight mark scatter points, a statistical chart of vision values, and vision values. The tester can choose to print them.

[0096] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An intelligent vision detection method, characterized in that: The following steps are involved: Preparation before testing: Information entry: Before the examination, the person being examined takes a facial photo and stores it in the database, and at the same time enters basic information including name, age and gender; Face verification: Perform face detection and face recognition on the inspected person, and compare the feature information with the face photos collected in the database. If the information is consistent, the next step can be started; Distance perception: The person being examined will be measured in real time, and the person will be prompted to adjust the examination distance until it is within the medically recommended distance range; Vision test: Obtaining the gaze point position of the person being inspected; Each sight mark is displayed randomly; Compare the coincidence between the gaze point position and the random position where the current sight mark appears. If the gaze point position appears within the sight mark area, it is considered that a sight mark has occurred, and record the inspection information when each sight mark occurs. Get the visual acuity result.

2. An intelligent vision detection method according to claim 1, characterized in that: When the person to be tested is an infant or young child with low cooperation, the pre-test preparation and vision test process also includes: playing animated images and audio to attract the attention of the person to be tested, while achieving the effect of reward and stimulating the person to be tested's interest in the test.

3. An intelligent vision detection method according to claim 2, characterized in that: During the distance sensing, the number of people at the inspected location is detected. If the number of people is greater than one, the vision test is only performed on the nearest inspected person, because the inspection of young infants and young children requires the accompaniment of their parents. At the same time, it can also prevent multiple people from accidentally entering the camera and causing the inspection to be interrupted.

4. The intelligent vision detection method according to claim 2, characterized in that: The specific process of obtaining the gaze point position of the inspected person is: capturing the dynamic features of the face and head of the inspected person in real time, completing the line of sight direction estimation, and calculating the gaze point position based on the actual size and resolution of the display device where the sight mark is located and the real-time distance between the inspected person and the display device, so as to achieve accurate projection of the viewpoint on the display device.

5. The intelligent vision detection method according to claim 1, characterized in that: The inspection information includes the position of the gaze point, the duration of gaze, and the real-time distance at the corresponding moment. Finally, the total number of gazes and the total duration of gazes during the current sight mark display time are counted and recorded. One sight mark corresponds to a complete record of inspection information.

6. The intelligent vision detection method according to claim 1, characterized in that: It also includes grading the vision test results: setting the initial level according to the visual ability of different age groups, and using a dichotomy method to automatically determine the level of the next test; If the current level fails, a lower level check will be conducted until a lower level is successfully passed, and this level is considered the vision level; If the current level is successfully passed, a higher level check will be carried out until a certain higher level fails, and the previous level of the level is considered to be the vision level.

7. An intelligent vision detection method according to claim 6, characterized in that: The visual acuity results include CPD visual acuity, Sellen decimal visual acuity and 5-point visual acuity.

8. The intelligent vision detection method according to claim 1, characterized in that: The vision test also includes: evaluating whether the cumulative duration of the visual behavior of the visual mark within each single display time period indicates that the gaze behavior is credible, and if it is credible, judging that the inspection of the current visual mark level has passed.

9. The intelligent vision detection method according to claim 7, characterized in that: After obtaining the visual acuity results, export the results of the visual acuity test and generate a report. The report includes a comparison chart of the fixation point and the visual mark scatter points, a visual acuity statistics chart, and a visual acuity value.

10. An intelligent vision detection system, applied to an intelligent vision detection method according to any one of claims 1 to 9, characterized in that: include: Dual-screen collaboration module: The secondary screen is the detection end screen, which displays all the function pages of vision screening, and the main screen is the screen of the inspected end, which displays different levels of visual marks and animations. The camera is located in the middle above the main screen; the secondary screen page has a separate area, which synchronizes the main screen page and the monitoring screen of the inspected person in real time to achieve dual-screen collaboration; Information entry module: Before the inspection, the inspected person takes a facial photo and stores it in the database, and at the same time enters basic information such as name, age, and gender; Face recognition module: performs face detection and face recognition on the subject, and compares the feature information with the face photos collected in the database. If the information is consistent, the next step can be started; Distance sensing module: used to measure the distance between the person being tested and the main screen in real time before and during the vision test; Attention-attracting module: This module can be used selectively. For infants and young children with low cooperation during the examination, animated images and audio can be played before and during the examination to attract the attention of the examinee, and at the same time, the effect of reward can be achieved to stimulate the examinee's interest in the examination. Pre-examination distance prompt submodule: If the information is successfully matched in the face recognition module, the distance of the person being examined will be measured in real time, and the person will be prompted to adjust the examination distance until it is within the medically recommended distance range, and then the vision examination can begin; Dynamic distance monitoring submodule: Real-time distance measurement during the whole process of vision examination, keeping real-time distance measurement, ensuring that only the person closest to the main screen is examined during the examination, because the examination of young infants and young children requires the company of their parents, and it can also prevent the situation where multiple people accidentally enter the camera and cause the examination to be interrupted; Viewpoint monitoring module: calculates the gaze point position of the inspected person; Viewpoint judgment module: Combined with the information of the dynamic distance monitoring submodule in the distance perception module, for the inspected person closest to the main screen, the viewpoint position calculated by the viewpoint monitoring module is compared with the random position where the current sight mark appears in real time. If the viewpoint position appears in the sight mark area, it is considered that a sight mark has occurred; Level judgment module: used to judge whether the current level of visual mark test has passed and whether it can enter a higher level of test; Credibility Assessment Submodule: It is used to assess whether the cumulative duration of the visual behavior of the visual mark within each single display period indicates that the gaze behavior is credible. The assessment standard can be set by the examiner on the secondary screen page and combined with clinical experience; Promotion judgment submodule: used to judge whether the current visual standard level inspection can be passed based on the results of the credibility assessment submodule; Conventional detection unit: Check from the lowest level step by step until the promotion judgment submodule determines that the current level has failed and locates the vision level to the previous level of the current level; Intelligent rapid detection unit: according to the visual ability of different age groups, the initial level is set and the next level of inspection is automatically determined by dichotomy; Test result conversion module: used to calculate the visual acuity value of the person being tested according to the test record value of the visual acuity level. The visual acuity results include CPD visual acuity value, Sellen decimal visual acuity value and 5-point visual acuity value; Report export module: used to export the results of vision screening. The content displayed in the report includes the comparison chart of fixation point and sight mark, the statistical chart of vision value, and the vision value. The tester can choose to print it.

Citation Information

Patent Citations

  • Method for viewing points detecting and apparatus thereof

    CN102473033A

  • Eye examination method and apparatus therefor

    CN110573061A

  • Sight line direction estimation method based on deep learning

    CN112800815A

  • Line-of-sight direction detection method and device

    CN115223231A

  • Visual function determination method and device based on eye movement tracking technology

    CN115944266A