A vision detection method based on projection and smart glasses

Abstract

The present invention relates to the field of electronic equipment, and in particular to a projection-based vision detection method and smart glasses. A projection-based vision detection method comprises: obtaining a fatigue score of a user to be detected; applying adjusted projection brightness, projection contrast and reaction limit time to perform vision detection; adjusting rest time based on the fatigue score, performing multiple vision detections, and performing weighted averaging on the obtained vision detection results to obtain a final vision detection value. The present invention evaluates the eye fatigue of the user to be detected by real-time monitoring of the user's eye reactions, such as eye movement, blinking frequency, pupil changes and other eye indicators, and adjusts the brightness and contrast of the projection content, as well as the set reaction limit time according to the user's real-time fatigue status, thereby realizing personalized and dynamic adjustment of the detection process; this personalized adjustment not only improves the comfort of the detection, but also ensures that accurate vision data can be obtained under different user states.

Description

Technical Field

[0001] The present invention relates to the field of electronic equipment, and in particular to a projection-based vision detection method and smart glasses. Background Art

[0002] In the field of traditional vision testing, accurately assessing individual vision status has always been a challenge, especially when the user has eye fatigue. Conventional vision testing methods rely on fixed vision charts and static testing environments, which may not accurately reflect the user's true vision level under different fatigue states. In addition, traditional vision chart testing often ignores individual differences in users, such as eye fatigue, reaction time and other factors, which may have a significant impact on vision test results.

[0003] Most of the current vision testing methods use a single testing standard and a fixed testing process, and lack the assessment and adaptive adjustment of the user's eye fatigue. These methods may not provide accurate test results when the user's eye fatigue is high, resulting in errors in vision assessment. At the same time, the accuracy of a single test is often affected by multiple factors, such as ambient light, personal fatigue status, and the accuracy of the testing equipment. These factors may lead to deviations in the test results, thereby affecting subsequent vision correction and eye health management decisions. Therefore, it is necessary to explore a new vision testing method that can comprehensively consider factors such as the user's eye fatigue and reaction time, and achieve accurate assessment of the user's vision by dynamically adjusting the test parameters and using advanced data processing technology. This can not only improve the scientificity and accuracy of vision testing, but also meet the needs of modern society for personalized medical services. Summary of the invention

[0004] The present invention monitors the user's eye reactions in real time, such as eye indicators such as eye movement, blinking frequency and pupil changes, evaluates the eye fatigue of the user to be tested, and adjusts the brightness and contrast of the projected content and the set reaction limit time according to the user's real-time fatigue status, thereby realizing personalized and dynamic adjustment of the detection process; this personalized adjustment not only improves the comfort of the detection, but also ensures that accurate vision data can be obtained under different user states.

[0005] A projection-based vision detection method, comprising:

[0006] S1: Evaluate the eye fatigue of the user to be tested and obtain a fatigue score;

[0007] S2: According to the obtained fatigue score, the brightness and contrast of the projected content are adjusted within a preset standard range. The higher the fatigue score, the lower the brightness and contrast of the projected content. According to the obtained fatigue score, the reaction limit time is selected within a preset reaction limit time range. The higher the fatigue score, the longer the reaction limit time.

[0008] Apply the selected projection brightness, projection contrast and reaction limit time to test the vision of the user to be tested, and calculate the average judgment time;

[0009] S3: According to the obtained fatigue score, select a rest time within the preset rest time range, the higher the fatigue score, the longer the rest time; repeat steps S1-S2 to obtain several vision test results, each vision test result includes a vision test value, a fatigue score and an average judgment time; the time interval between each two executions is the selected rest time; based on the fatigue score and the average judgment time of the user to be tested in each vision test result, perform a weighted average on all obtained vision test values, the lower the fatigue score and the average judgment time, the higher the weight, and the weight ratio used for weighted averaging is obtained by the weight ratio calculation formula; the 5-point record value closest to the obtained weighted average is the final vision test value.

[0010] Preferably, the specific operation of performing weighted averaging on all acquired vision test values ​​is:

[0011] Get n vision test results, including n vision test values ​​x i , fatigue score F i and the average judgment time T i , i=1,2,…,n;

[0012] Using the formula Calculate the weighted average value X; where ω i is the visual acuity test value x i The weight ratio, weight ratio ω i The calculation method is as follows:

[0013] Using the formula Calculate the visual acuity test value x i The weight value W i ; Where a>0 and b>0, a and b are used to control the fatigue F to the weight value W i The degree of influence; c>0, c is used to adjust the average judgment time T on the weight value W i The degree of influence; then using the formula Calculate the visual acuity test value x i The weight ratio ω i ;

[0014] Preferably, the specific operation of evaluating the eye fatigue of the user to be detected and obtaining the fatigue score is as follows:

[0015] S11: setting a standardized projection content for the user to be detected to watch, and monitoring the user's eye movement speed and frequency in real time through an eye tracking device; comparing the obtained eye movement speed and eye movement frequency with the benchmark movement speed and benchmark movement frequency in the pre-acquired group benchmark data, and scoring the degree of deviation of the eye movement speed and eye movement frequency based on the difference obtained by the comparison;

[0016] S12: using a camera to monitor the blinking frequency and blinking duration of the user to be detected, and at the same time checking the red blood streaks in the eyes of the user to be detected; comparing the acquired blinking frequency, blinking duration and red blood streaks with the benchmark blinking frequency, benchmark blinking duration and benchmark red blood streaks in the pre-acquired group benchmark data, and scoring the deviation degree of the blinking frequency, blinking duration and red blood streaks respectively based on the difference obtained by the comparison;

[0017] S13: Detecting pupil reaction speed and diameter change amplitude of the pupil to light by using an infrared pupil tracking device; comparing the obtained pupil reaction speed and pupil diameter change amplitude with the benchmark reaction speed and benchmark diameter change amplitude in the pre-acquired group benchmark data, and scoring the degree of deviation of the pupil reaction speed and pupil diameter change amplitude respectively based on the difference obtained by the comparison;

[0018] S14: The obtained eye movement speed, eye movement frequency, blinking frequency, blinking duration, red bloodshot condition, pupil reaction speed and pupil diameter change amplitude are averaged to obtain a fatigue score.

[0019] Preferably, the specific operation of obtaining the benchmark data is:

[0020] A1: Establish an eye dataset, which includes benchmark data obtained by monitoring people of different ages and genders watching standardized projection content under normal conditions. The benchmark data includes eye movement speed, eye movement frequency, blinking frequency, blinking duration, red bloodshot conditions, pupil reaction speed, and pupil diameter change range; calculate the average value of each indicator in all individual benchmark data to obtain group benchmark data;

[0021] A2: Establish an eye database, repeat step A1 in each preset update cycle, and use the eye data set of the current update cycle to replace the eye data set of the previous update cycle to obtain the group benchmark data of the current update cycle.

[0022] Preferably, the specific operation of performing vision testing on the user to be tested includes:

[0023] The projection content is a standard visual acuity logarithmic chart. The user to be tested shall judge the visual marks row by row from top to bottom. If the number of correct judgments for a row of visual marks is greater than half of the total number of visual marks in the current row, the user shall continue to judge the visual marks in the next row until the number of correct judgments for a row of visual marks is less than or equal to half of the total number of visual marks in the current row. The visual marks in the previous row of the current row shall be taken as the minimum row of visual marks for the user to be tested. The 5-point record value corresponding to the minimum row of visual marks is the visual acuity test value of the user to be tested.

[0024] Among them, the specific operation of distinguishing the sight mark is:

[0025] For any visual mark, the judgment time of the user to be tested is recorded. If the judgment result of the user to be tested is correct and the judgment time is within the reaction limit, it is considered to be a correct judgment; otherwise, it is considered to be an incorrect judgment;

[0026] At the same time, the camera is used to monitor the eye state of the user to be detected in real time to determine whether there is squinting behavior. If squinting behavior is present, it is directly regarded as an error in judgment.

[0027] Preferably, the lowest reaction limit time of the reaction limit time range is obtained by a swarm optimization algorithm, and the highest reaction limit time of the reaction limit time is twice the lowest reaction limit time.

[0028] Preferably, the swarm optimization algorithm for obtaining the minimum reaction limit time is a particle swarm optimization algorithm, and the specific operations are:

[0029] B1: Randomly generate a particle swarm, and the position of each particle represents a candidate limit time;

[0030] B2: For any particle, initialize the velocity of the particle and take the current position as the individual optimal position of the particle; use the formula Calculate the fitness value F of the particle, where D represents the accuracy of vision detection using the current candidate limit time, L represents the duration of the current candidate limit time; β1 and β2 are the weights of D and L respectively;

[0031] B3: Traverse all particles and take the position corresponding to the particle with the highest fitness value F as the global optimal position;

[0032] B4: For any particle, update the speed and position of the particle, and recalculate the fitness value; if the updated fitness value is higher than the fitness value of the individual best position of the particle, update the individual best position of the particle to the current position;

[0033] B5: Traverse all particles. If the fitness value of the individual best position of any particle is higher than the fitness value of the global best position, the global best position is updated to the individual best position of the particle.

[0034] B6: Repeat steps B4 to B5 until the maximum number of iterations is reached. The global best position obtained is the optimal limit time.

[0035] Preferably, a projection-based vision detection smart glasses are applied to any of the above-mentioned projection-based vision detection methods.

[0036] The present invention has the following advantages:

[0037] 1. The present invention monitors the user's eye reactions in real time, such as eye movements, blinking frequency, pupil changes and other eye indicators, evaluates the eye fatigue of the user to be tested, and adjusts the brightness and contrast of the projected content and the set reaction limit time according to the user's real-time fatigue status, thereby realizing personalized and dynamic adjustment of the detection process; this personalized adjustment not only improves the comfort of the detection, but also ensures that accurate vision data can be obtained under different user states.

[0038] 2. The present invention adopts the method of multiple tests and weighted averaging to improve the accuracy and reliability of vision test results; each test result is weighted based on the user's fatigue score and average judgment time, so as to ensure that the test results obtained when the user is in the best state have a higher influence; this method effectively reduces the impact of accidental errors and random fluctuations on the test results, especially when the user may have vision fluctuations; finally, by comprehensively considering all test results, the present invention can provide a stable and reliable vision assessment, provide users with more accurate vision information, and provide solid data support for subsequent vision correction and eye health management. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A schematic flow chart of a projection-based vision detection method used in an embodiment of the present invention.

[0040] Figure 2 The figure is a flow chart of a method for obtaining a fatigue degree score according to an embodiment of the present invention. DETAILED DESCRIPTION

[0041] In order to enable persons skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0042] Embodiment, a method for visual acuity detection based on projection, such as Figure 1-Figure 2 As shown, including:

[0043] S1: Evaluate the eye fatigue of the user to be tested and obtain a fatigue score. By evaluating eye fatigue, the system can collect important data about the user's current visual comfort and stress level. In a fatigued state, the eye's sensitivity may decrease, making it difficult for the user to accurately identify the visual mark during vision testing. Fatigue may also affect the pupil's reaction speed and accuracy to light changes, which may affect the clarity of the visual mark during vision testing. On the other hand, eye fatigue may be accompanied by discomfort such as dry eyes, itching or pain. These discomforts will distract the user and affect the accuracy of vision testing. At the same time, the speed at which the user's brain processes visual information may slow down, resulting in a decrease in the speed of identifying and responding to the visual mark during vision testing. These data are crucial to understanding the user's visual ability and how to adjust the testing process to ensure that the test results are more accurate and user-friendly.

[0044] S2: According to the obtained fatigue score, adjust the brightness and contrast of the projected content within the preset standard range. The higher the fatigue score, the lower the brightness and contrast of the projected content, so as to reduce stimulation to the eyes and make the visual detection process gentler; according to the obtained fatigue score, select the reaction limit time within the preset reaction limit time range. The higher the fatigue score, the longer the reaction limit time, so as to give the user more time to react; the purpose of adjusting the brightness and contrast is to adapt to the user's visual comfort and avoid aggravating eye fatigue due to excessive brightness or contrast, which affects the accuracy of the detection results; the purpose of extending the reaction limit time is to take into account that the user may need more time to identify and react to the visual mark in a fatigued state, thereby reducing the detection error caused by time pressure and improving the accuracy of the detection and the user's detection experience;

[0045] For the standard range of projection content brightness, according to the lighting requirements of the national standard GB 11533-2011 "Standard Logarithmic Vision Chart", the brightness of the white background of the vision chart measured by a brightness meter should be no less than 200cd / m 2 (candela per square meter), that is, 200 candela per square meter of brightness; since there are standard values, we will not elaborate on them here. The highest brightness in the standard range only needs to be guaranteed not to be dazzling; and the standard range of contrast is the same, just need to be guaranteed to be within a suitable range;

[0046] The selected projection brightness, projection contrast and reaction limit time are applied to test the vision of the user to be tested. The vision test adopts the national standard GB11533-2011 standard logarithmic vision chart, and the average judgment time is calculated at the same time;

[0047] S3: According to the fatigue score obtained, select a rest time within the preset rest time range. The higher the fatigue score, the longer the rest time, to ensure that the user's eyes get enough rest. Reasonable arrangement of rest time can help reduce the impact of eye fatigue on vision test results, and ensure that the user can maintain a good visual state in each test; repeat steps S1-S2 to obtain several vision test results, each vision test result includes a vision test value, a fatigue score and an average judgment time; the time interval between each two executions is the selected rest time. Through multiple tests, the user's vision data under different visual states can be collected to provide more comprehensive information for subsequent data processing; based on each vision test, the user's vision data under different visual states can be collected to provide more comprehensive information for subsequent data processing. The fatigue score and average judgment time of the user to be tested in the test results are weighted averaged for all the obtained vision test values. The lower the fatigue score and the average judgment time, the higher the weight, and the weight ratio used for weighted averaging is obtained by the weight ratio calculation formula, because these data can reflect the user's visual ability and reaction speed at each test. If the fatigue score and the average judgment time are low, that is, the user's visual state is good, the weighted average can more accurately reflect the user's true vision level and reduce the deviation of single test results caused by eye fatigue or accidental factors; the 5-point record value closest to the obtained weighted average is the final vision test value, where the 5-point record value (Fractional LogMAR Chart) is a system for measuring vision, which provides a standardized and detailed method to evaluate and record vision levels.

[0048] The specific operation of weighted averaging all acquired vision test values ​​is as follows:

[0049] Get n vision test results, including n vision test values ​​x i , fatigue score F i and the average judgment time T i , i=1,2,…,n;

[0050] Using the formula Calculate the weighted average value X; where ω i is the visual acuity test value x i The weight ratio, weight ratio ω i The calculation method is as follows:

[0051] Using the formula Calculate the visual acuity test value x i The weight value W i ; Where a>0 and b>0, a and b are used to control the fatigue F to the weight value W ia controls the speed of change of the reduction rate, and b determines the turning point of the acceleration of the reduction rate, that is, when the fatigue score exceeds b, the weight value decreases significantly faster; c>0, c is used to adjust the average judgment time T to the weight value W i The degree of influence; then using the formula Calculate the visual acuity test value x i The weight ratio ω i ;

[0052] This scheme takes into account that the effects of fatigue and average judgment time on detection accuracy do not change linearly. For example, when the user to be detected is slightly fatigued, although the eyes may have begun to feel some fatigue, the brain and visual system can usually still compensate for these effects well and maintain a high level of vision detection accuracy. Therefore, the difference between the vision detection results under mild fatigue and normal conditions may be small. As fatigue increases to moderate, the visual system begins to be more obviously affected, such as slower eye movement, increased blinking frequency, decreased attention, etc. These changes will have a more obvious negative impact on vision detection accuracy, but may not be drastic changes. In a state of severe fatigue, the decline in visual and cognitive functions may become very obvious. For example, the eyes may be more difficult to focus, reaction time may be prolonged, judgment may be reduced, and even temporary blurred vision may occur. At this time, the accuracy of vision detection may drop significantly and the error rate may increase significantly. Therefore, the effect of fatigue score on weight value is: the weight value decreases as the fatigue score increases, and the rate of decrease of weight value becomes faster and faster.

[0053] The average judgment time has a milder effect on the weight. Even if the user's reaction time is longer, the weight decreases more slowly. This takes into account the possibility that the user's reaction time may be prolonged due to careful identification, and avoids excessive punishment for thoughtful judgment.

[0054] Evaluate the eye fatigue of the user to be tested and obtain the specific operation of the fatigue score:

[0055] S11: Set a standardized projection content. This content does not need to be too complicated and non-irritating. It only needs to be enough to attract attention and induce eye movement and pupil reaction. The playback time of the evaluation content should be moderate, usually a few minutes, so as to collect enough data without further aggravating the user's eye fatigue. The user to be tested watches it, and at the same time, the eye movement speed and frequency of the user's eyeballs are monitored in real time through an eye tracking device. When the eyes are tired, the eye movement may become slow and irregular. The obtained eye movement speed and eye movement frequency are compared with the benchmark movement speed and benchmark movement frequency in the pre-acquired group benchmark data. Based on the difference obtained by the comparison, the deviation degree of the eye movement speed and eye movement frequency is scored respectively;

[0056] S12: Using a camera to monitor the blinking frequency and blinking duration of the user to be detected, and at the same time checking the red blood streaks in the eyes of the user to be detected. When the eyes are tired, the blinking frequency usually increases, because the eyes are dry and need to be lubricated more frequently; at the same time, the duration of each blink may become longer; on the other hand, obvious red blood streaks may appear in the eyes, which is the result of dilation of blood vessels in the eyes; this change can be detected by a high-definition camera; the acquired blinking frequency, blinking duration and red blood streaks are compared with the baseline blinking frequency, baseline blinking duration and baseline red blood streaks in the pre-acquired group baseline data, and based on the difference obtained by the comparison, the deviation degree of the blinking frequency, blinking duration and red blood streaks is scored respectively;

[0057] S13: Detect the pupil's reaction speed to light and the change in diameter by using an infrared pupil tracking device. When fatigued, the pupil's reaction to light becomes slow and the change in pupil diameter may slow down. At the same time, the pupil may show unstable micro-tremors when fatigued, which is also an indication of fatigue. Compare the obtained pupil reaction speed and pupil diameter change with the baseline reaction speed and baseline diameter change in the pre-acquired group baseline data. Based on the difference obtained by the comparison, score the deviation of the pupil reaction speed and pupil diameter change respectively.

[0058] The above scores are all scored in the range of 0-10. For example, when the blink duration is within 10% of the baseline blink duration, the score is 1, and when the deviation is within 20%, the score is 2, and so on.

[0059] S14: The obtained eye movement speed, eye movement frequency, blinking frequency, blinking duration, red bloodshot condition, pupil reaction speed and pupil diameter change amplitude are averaged to obtain a fatigue score.

[0060] Specific operations for obtaining benchmark data:

[0061] A1: Establish an eye dataset, which includes benchmark data obtained by monitoring people of different ages and genders watching standardized projection content under normal conditions. By collecting this data, a benchmark reflecting the eye activity pattern under normal visual conditions can be established; these benchmark data are crucial for subsequent eye fatigue assessments because they provide a standard reference for normal visual function; benchmark data include eye movement speed, eye movement frequency, blinking frequency, blinking duration, red bloodshot conditions, pupil reaction speed, and pupil diameter change amplitude; calculate the average of each indicator in all individual benchmark data to obtain group benchmark data, which provides a standardized reference point for assessing individual eye fatigue; individual eye response data will be compared with these averages to determine their fatigue score;

[0062] A2: Establish an eye database to store and manage group benchmark data collected at different time points. Repeat step A1 in each preset update cycle, and use the eye dataset of the current update cycle to replace the eye dataset of the previous update cycle to obtain the group benchmark data of the current update cycle. Therefore, this database is dynamically updated and can continuously optimize and adjust the benchmark data over time to adapt to changes in the visual characteristics of the population. By replacing old data, it can ensure that the vision detection method is always based on the latest group visual characteristics, thereby improving the relevance and accuracy of the detection.

[0063] The specific operations for performing vision testing on the user to be tested include:

[0064] The projected content is a standard visual acuity logarithmic chart, which is a common vision testing tool. The size of the visual symbols increases or decreases in a geometric series. The user to be tested judges the visual symbols row by row from top to bottom. If the number of correct judgments for a row of visual symbols is greater than half of the total number of visual symbols in the current row, the visual symbols in the next row will be judged. This standard ensures the objectivity of the test results. Only when the user shows a stable ability to recognize the visual symbols will the test of the next row be continued; until the number of correct judgments for a row of visual symbols is less than or equal to half of the total number of visual symbols in the current row, it means that the user has reached the limit of vision and cannot recognize smaller visual symbols. The test results at this time will be used to determine the user's vision level, and the visual symbols in the previous row of the current row will be used as the minimum row of visual symbols for the user to be tested. The 5-point record value corresponding to the minimum row of visual symbols is the vision test value of the user to be tested. This decreasing process helps to determine the precise lower limit of the user's vision, that is, the minimum size of the visual symbol that the user can stably recognize.

[0065] Specific operations for identifying visual marks:

[0066] For any visual mark, the judgment time of the user to be tested is recorded. This is done by timing the time it takes the user to see the visual mark and react. The recording of judgment time helps to evaluate the user's visual processing speed and concentration, and is an important indicator for evaluating the user's visual reaction ability. If the judgment result of the user to be tested is correct and the judgment time is within the reaction limit, the judgment is considered correct; otherwise, it is considered wrong. This dual standard ensures that the test result is not only based on the user's visual recognition ability, but also takes into account the user's visual reaction speed, thereby more comprehensively evaluating the user's vision condition;

[0067] At the same time, the camera is used to monitor the eye state of the user to be tested in real time to determine whether there is squinting behavior. Squinting behavior is usually a natural reaction to reduce the amount of light entering the eyes or improve visual clarity, but in standardized vision tests, this may indicate that the user cannot accurately identify the visual mark under normal vision conditions. Therefore, treating it as a wrong judgment helps maintain the rigor and standardization of the test. If squinting behavior is present, it is directly regarded as a judgment error. By monitoring the user's natural eye reactions, the user's visual ability in an unconscious state can be more realistically evaluated, avoiding the impact of eye fatigue or unnatural eye posture on the test results.

[0068] The lowest reaction limit time of the reaction limit time range is obtained by a swarm optimization algorithm, which is a type of algorithm that imitates the behavior of biological groups in nature and is used to solve optimization problems; these algorithms usually include a set of candidate solutions, which are continuously updated during the iteration process in order to find the optimal or approximately optimal solution to the problem; the highest reaction limit time of the reaction limit time is twice the lowest reaction limit time.

[0069] The swarm optimization algorithm for obtaining the minimum reaction time limit is the particle swarm optimization algorithm. The reason for using the particle swarm optimization algorithm (PSO) is mainly to determine the optimal reaction time limit, which is a key parameter to ensure the accuracy of vision detection. As a swarm intelligence algorithm that simulates the foraging behavior of bird flocks, PSO has the advantages of simple implementation, few parameters, and strong adaptability, and can efficiently search for the optimal solution in a complex environment. The advantage of using PSO is that it can quickly converge and find an approximate global optimal solution, while avoiding the problem that the traditional gradient descent algorithm may fall into the local optimum. In addition, PSO does not require the gradient information of the problem and is also applicable to nonlinear and non-convex problems, which makes it very suitable for the optimization of the reaction time limit in this solution to improve the accuracy and efficiency of vision detection. The reaction time limit obtained by PSO optimization can balance the accuracy of detection and the reaction speed of the user, thereby improving the overall detection experience and the reliability of the results. Specific operations:

[0070] B1: Randomly generate a particle swarm. The position of each particle represents a candidate limit time. By randomizing the initial solution, the possibility of finding the global optimal solution is increased and the algorithm is prevented from falling into the local optimum.

[0071] B2: For any particle, initialize the velocity of the particle and take the current position as the individual optimal position of the particle; use the formula Calculate the fitness value F of the particle, where D represents the accuracy of vision detection using the current candidate limit time, and L represents the duration of the current candidate limit time; β1 and β2 are the weights of D and L respectively; the fitness function is used to evaluate the quality of each candidate solution. By assigning weights to accuracy and duration, the importance of the two in the optimization process can be balanced;

[0072] B3: Traverse all particles and take the position corresponding to the particle with the highest fitness value F as the global best position; the global best position guides the direction of the group search, and all particles will be updated according to this direction in order to find a better solution;

[0073] B4: For any particle, update the speed and position of the particle, and recalculate the fitness value; if the updated fitness value is higher than the fitness value of the individual best position of the particle, update the individual best position of the particle to the current position;

[0074] B5: Traverse all particles. If the fitness value of the individual best position of any particle is higher than the fitness value of the global best position, the global best position is updated to the individual best position of the particle.

[0075] B6: Repeat steps B4-B5 until the maximum number of iterations is reached. The global best position obtained is the optimal limit time. Through iterative search, the particle swarm optimization algorithm can converge to the optimal or approximately optimal solution after multiple generations of search, ensuring the stability and reliability of the algorithm.

[0076] A projection-based vision detection smart glasses, wherein the smart glasses are applied to any of the above-mentioned projection-based vision detection methods.

[0077] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.

Claims

1. A projection-based vision detection method, characterized in that: include: S1: Evaluate the eye fatigue of the user to be tested and obtain a fatigue score; S2: According to the obtained fatigue score, the brightness and contrast of the projected content are adjusted within a preset standard range. The higher the fatigue score, the lower the brightness and contrast of the projected content. According to the obtained fatigue score, the reaction limit time is selected within a preset reaction limit time range. The higher the fatigue score, the longer the reaction limit time. Apply the selected projection brightness, projection contrast and reaction limit time to test the vision of the user to be tested, and calculate the average judgment time; S3: According to the fatigue score obtained, a rest time is selected within a preset rest time range, and the higher the fatigue score, the longer the rest time; Step S1-Step S2 are repeated to obtain a number of vision test results, each of which includes a vision test value, a fatigue score and an average judgment time; the time interval between two executions is the selected rest time; based on the fatigue score and the average judgment time of the user to be tested in each vision test result, all the obtained vision test values ​​are weighted averaged, and the lower the fatigue score and the average judgment time, the higher the weight, and the weight ratio used for weighted averaging is obtained by the weight ratio calculation formula; the 5-point record value closest to the obtained weighted average value is the final vision test value; The specific operation of weighted averaging all acquired vision test values ​​is as follows: Get n vision test results, including n vision test values ​​x i , fatigue score F i and the average judgment time T i , i = 1, 2, ..., n; Using the formula Calculate the weighted average value X; where ω i is the visual acuity test value x i The weight ratio, weight ratio ω i The calculation method is as follows: Using the formula Calculate the visual acuity test value x i The weight value W i ; Where a>0 and b>0, a and b are used to control the fatigue F to the weight value W i The degree of influence; c>0, c is used to adjust the average judgment time T on the weight value W i The degree of influence; then using the formula Calculate the visual acuity test value x i The weight ratio ωi.

2. A projection-based vision detection method according to claim 1, characterized in that: Evaluate the eye fatigue of the user to be tested and obtain the specific operation of the fatigue score: S11: setting a standardized projection content for the user to be detected to watch, and monitoring the user's eye movement speed and frequency in real time through an eye tracking device; comparing the obtained eye movement speed and eye movement frequency with the benchmark movement speed and benchmark movement frequency in the pre-acquired group benchmark data, and scoring the degree of deviation of the eye movement speed and eye movement frequency based on the difference obtained by the comparison; S12: using a camera to monitor the blinking frequency and blinking duration of the user to be detected, and at the same time checking the red blood streaks in the eyes of the user to be detected; comparing the acquired blinking frequency, blinking duration and red blood streaks with the benchmark blinking frequency, benchmark blinking duration and benchmark red blood streaks in the pre-acquired group benchmark data, and scoring the deviation degree of the blinking frequency, blinking duration and red blood streaks respectively based on the difference obtained by the comparison; S13: Detecting pupil reaction speed and diameter change amplitude of the pupil to light by using an infrared pupil tracking device; comparing the obtained pupil reaction speed and pupil diameter change amplitude with the benchmark reaction speed and benchmark diameter change amplitude in the pre-acquired group benchmark data, and scoring the degree of deviation of the pupil reaction speed and pupil diameter change amplitude respectively based on the difference obtained by the comparison; S14: The obtained eye movement speed, eye movement frequency, blinking frequency, blinking duration, red bloodshot condition, pupil reaction speed and pupil diameter change amplitude are averaged to obtain a fatigue score.

3. A projection-based vision detection method according to claim 2, characterized in that: Specific operations for obtaining benchmark data: A1: Establish an eye dataset, which includes benchmark data obtained by monitoring people of different ages and genders watching standardized projection content under normal conditions. The benchmark data includes eye movement speed, eye movement frequency, blinking frequency, blinking duration, red bloodshot conditions, pupil reaction speed, and pupil diameter change range; calculate the average value of each indicator in all individual benchmark data to obtain group benchmark data; A2: Establish an eye database, repeat step A1 in each preset update cycle, and use the eye data set of the current update cycle to replace the eye data set of the previous update cycle to obtain the group benchmark data of the current update cycle.

4. A projection-based vision detection method according to claim 3, characterized in that: The specific operations for performing vision testing on the user to be tested include: The projection content is a standard visual acuity logarithmic chart. The user to be tested shall judge the visual marks row by row from top to bottom. If the number of correct judgments for a row of visual marks is greater than half of the total number of visual marks in the current row, the user shall continue to judge the visual marks in the next row until the number of correct judgments for a row of visual marks is less than or equal to half of the total number of visual marks in the current row. The visual marks in the previous row of the current row shall be taken as the minimum row of visual marks for the user to be tested. The 5-point record value corresponding to the minimum row of visual marks is the visual acuity test value of the user to be tested. Among them, the specific operation of distinguishing the sight mark is: For any visual mark, the judgment time of the user to be tested is recorded. If the judgment result of the user to be tested is correct and the judgment time is within the reaction limit time, it is considered to be a correct judgment; otherwise, it is considered to be an incorrect judgment; At the same time, the camera is used to monitor the eye state of the user to be detected in real time to determine whether there is squinting behavior. If squinting behavior is present, it is directly regarded as an error in judgment.

5. A projection-based vision detection method according to claim 4, characterized in that: The lowest reaction limit time of the reaction limit time range is obtained by the group optimization algorithm, and the highest reaction limit time of the reaction limit time is twice the lowest reaction limit time.

6. A projection-based vision detection method according to claim 5, characterized in that: The swarm optimization algorithm for obtaining the minimum reaction limit time is the particle swarm optimization algorithm. The specific operations are: B1: Randomly generate a particle swarm, and the position of each particle represents a candidate limit time; B2: For any particle, initialize the velocity of the particle and take the current position as the individual optimal position of the particle; use the formula Calculate the fitness value F of the particle, where D represents the accuracy of vision detection using the current candidate limit time, L represents the duration of the current candidate limit time; β1 and β2 are the weights of D and L respectively; B3: Traverse all particles and take the position corresponding to the particle with the highest fitness value F as the global optimal position; B4: For any particle, update the speed and position of the particle, and recalculate the fitness value; if the updated fitness value is higher than the fitness value of the individual best position of the particle, then update the individual best position of the particle to the current position; B5: Traverse all particles. If the fitness value of the individual best position of any particle is higher than the fitness value of the global best position, the global best position is updated to the individual best position of the particle. B6: Repeat steps B4 to B5 until the maximum number of iterations is reached. The global best position obtained is the optimal limit time.

7. A projection-based vision detection smart glasses, characterized in that: The smart glasses are applied to the projection-based vision detection method described in any one of claims 1 to 6 above.

Images