Multi-modal interactive color vision screening system and method

Abstract

The invention discloses a multi-modal interactive color vision screening system and method, and relates to the technical field of color vision screening, and the key points of the technical scheme comprise the following steps: obtaining a multi-dimensional scene reference parameter, a student sight tracking calibration parameter and a parameter attribute type of color vision screening; capturing sight track real-time monitoring data and screening scene dynamic adjustment data of a target student; performing comparative analysis on the matched calibration item and a calibration standard item corresponding to the matched calibration item to obtain a calibration abnormal item in the matched calibration item; analyzing according to the missing calibration item to obtain a first calibration abnormal coefficient of color vision screening in the target calibration monitoring domain; analyzing according to the calibration anomaly item to obtain a second calibration anomaly coefficient of color vision screening in the target calibration monitoring domain; obtaining a comprehensive calibration abnormal coefficient according to the first calibration abnormal coefficient and the second calibration abnormal coefficient; and outputting a color vision screening report according to the comprehensive calibration abnormal coefficient. The method has the effect of improving the efficiency and generalizability of color vision screening.

Description

Technical Field

[0001] This invention relates to the field of color vision screening technology, and more specifically, to a multimodal interactive color vision screening system and method. Background Technology

[0002] Early color vision screening relied heavily on static color charts, which could only provide fixed color stimuli and could not adapt to different lighting conditions, display devices, and other environmental factors. This could lead to discrepancies in screening results for the same student in different environments. Furthermore, traditional methods lacked monitoring of the test subject's state. For example, if a student's gaze was scattered or their focus was inaccurate during screening, the screening results would be affected by their subjective state, which could not be identified and corrected in a timely manner.

[0003] With the increasing demand for color vision screening in the education and medical fields, single-dimensional screening methods are no longer sufficient. Different students have different visual perception abilities and attention states, and a fixed screening process cannot match individual characteristics; fluctuations in environmental parameters of the screening scenario will directly interfere with the color presentation effect, thus affecting the accuracy of the screening. Summary of the Invention

[0004] In view of the shortcomings of the existing technology, the purpose of this invention is to provide a multimodal interactive color vision screening system and method.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A multimodal interactive color vision screening method, comprising the following steps: Acquire multi-dimensional scene baseline parameters, student gaze tracking calibration parameters, and parameter attribute types for color vision screening; capture real-time monitoring data of the target student's gaze trajectory and dynamic adjustment data of the screening scene; Construct a scene gaze calibration monitoring domain based on parameter attribute types, and generate calibration standard items for the scene gaze calibration monitoring domain based on multi-dimensional scene benchmark parameters and student gaze tracking calibration parameters; Based on real-time monitoring data of eye trajectory and dynamic adjustment data of screening scenario, calibration and adaptation color vision monitoring items are generated. The monitoring attribute type of calibration and adaptation color vision monitoring items is compared and analyzed with the scene eye calibration monitoring domain to obtain the target calibration monitoring domain corresponding to the calibration and adaptation color vision monitoring items. The calibration-adapted color vision monitoring items are compared and analyzed with the calibration standard items of the target calibration monitoring domain to obtain the matching calibration items and the missing calibration items; the matching calibration items are compared and analyzed with the corresponding calibration standard items to obtain the calibration anomalies in the matching calibration items. The first calibration anomaly coefficient of color vision screening in the target calibration monitoring domain was obtained based on the analysis of missing calibration items; the second calibration anomaly coefficient of color vision screening in the target calibration monitoring domain was obtained based on the analysis of calibration anomalies. The comprehensive calibration anomaly coefficient is obtained based on the first calibration anomaly coefficient and the second calibration anomaly coefficient; the color vision screening report is output based on the comprehensive calibration anomaly coefficient.

[0006] Preferably, the multi-dimensional scene reference parameters include color stimulus source parameters, ambient light intensity reference parameters, display interface adaptation parameters, and spatial layout reference parameters; The student eye tracking calibration parameters include eye focus parameters, eye movement rate parameters, eye dwell time parameters, and eye deviation threshold parameters. Among them, the multi-dimensional scene benchmark parameters and student gaze tracking calibration parameters both include the corresponding benchmark update cycle; The parameter attribute types include scene baseline type and line-of-sight calibration type.

[0007] Preferably, the calibration and adaptation color vision monitoring items are generated based on real-time monitoring data of the gaze trajectory and dynamic adjustment data of the screening scenario, specifically including the following steps: Based on parameter attribute types, scene baseline monitoring parameters and line of sight calibration monitoring parameters are extracted from real-time line of sight trajectory monitoring data and dynamic adjustment data of screening scenarios. Generate scene benchmark monitoring items based on scene benchmark monitoring parameters; Generate line-of-sight calibration monitoring items based on line-of-sight calibration monitoring parameters; The calibration and adaptation color vision monitoring items for color vision screening are composed of scene baseline monitoring items and line-of-sight calibration monitoring items.

[0008] Preferably, the monitoring attribute type of the calibration-adapted color vision monitoring item is compared and analyzed with the scene gaze calibration monitoring domain to obtain the target calibration monitoring domain corresponding to the calibration-adapted color vision monitoring item. This specifically includes the following steps: Compare the attribute types of the monitoring attribute types corresponding to the calibration and adaptation color vision monitoring items with the parameter attribute types corresponding to the scene vision calibration monitoring domain; The scene gaze calibration monitoring field with the same parameter attribute type as the monitoring attribute type is marked as the target calibration monitoring field corresponding to the calibration adaptation color vision monitoring item.

[0009] Preferably, the calibration-adapted color vision monitoring items are compared and analyzed with the calibration standard items of the target calibration monitoring domain to obtain matching calibration items and missing calibration items. This specifically includes the following steps: Obtain the benchmark update cycle and historical calibration data corresponding to the calibration standard items in the target calibration monitoring domain; The calibration standard items to be calibrated are obtained based on the benchmark update cycle and historical calibration data; The calibration-adapted color vision monitoring items are compared with the calibration standard items of the target calibration monitoring domain. Mark the calibration-adaptive color vision monitoring items that match the calibration standard items in the target calibration monitoring domain as matching calibration items; Mark the standard items to be calibrated in the target calibration monitoring domain that do not match any of the calibration-adapted color vision monitoring items as missing calibration items.

[0010] Preferably, the first calibration anomaly coefficient of color vision screening in the target calibration monitoring domain is obtained based on the analysis of missing calibration items, specifically including the following steps: Obtain the number of missing calibration items and the number of parameters to be calibrated in the target calibration monitoring domain; The missing percentage of the missing calibration items in the target calibration monitoring domain is obtained based on the number of missing items and the number of items to be calibrated. The missing percentage coefficient of the missing calibration items is obtained based on the missing percentage value. The missing calibration period value of the missing calibration item is obtained based on the benchmark update cycle of the missing calibration item and historical calibration data, and the missing calibration period coefficient of the missing calibration item is obtained based on the missing calibration period value. The missing anomaly coefficient of the missing calibration item is obtained based on the missing percentage coefficient and the missing period coefficient. The missing anomaly coefficients of the missing calibration items in the target calibration monitoring domain are all accumulated to obtain the first calibration anomaly coefficient of color vision screening in the target calibration monitoring domain.

[0011] Preferably, the second calibration anomaly coefficient for color vision screening in the target calibration monitoring domain is obtained based on calibration anomaly analysis, specifically including the following steps: The parameter anomaly coefficient is obtained by comparing the calibration monitoring parameters of the matching calibration item with the standard parameters of the corresponding calibration standard item. The parameter anomaly coefficients corresponding to the calibration anomalies in the target calibration monitoring domain are summed to obtain the second calibration anomaly coefficient of color vision screening in the target calibration monitoring domain.

[0012] Preferably, the parameter anomaly coefficient is obtained by comparing the calibration monitoring parameters of the matching calibration item with the standard parameters of the corresponding calibration standard item, specifically including the following steps: If the calibration monitoring parameters of the matched calibration item meet the standard parameters, then the matched calibration item is judged to be a normal calibration item; If the calibration monitoring parameters of the matched calibration item do not meet the standard parameters, the matched calibration item is judged to be a calibration anomaly. Based on the calibration monitoring parameters of the calibration anomaly and the standard parameters of the corresponding calibration standard, the parameter anomaly coefficient corresponding to the calibration anomaly is obtained.

[0013] Preferably, the comprehensive calibration anomaly coefficient is obtained based on the first calibration anomaly coefficient and the second calibration anomaly coefficient, specifically including the following steps: The calibration anomaly coefficient of the color vision screening in the target calibration monitoring domain is obtained based on the first calibration anomaly coefficient and the second calibration anomaly coefficient. Set the monitoring domain weights corresponding to the target calibration monitoring domain according to the parameter attribute types corresponding to the target calibration monitoring domain; The monitoring domain calibration anomaly coefficient of the target calibration monitoring domain is obtained by assigning weights to the monitoring domain weights. The anomaly coefficients of the monitoring fields of each scene's gaze calibration monitoring field are summed to obtain the comprehensive calibration anomaly coefficient for color vision screening.

[0014] A multimodal interactive color vision screening system, comprising: Acquisition Module: Acquires multi-dimensional scene baseline parameters for color vision screening, student gaze tracking calibration parameters, and parameter attribute types; captures real-time monitoring data of the target student's gaze trajectory and dynamic adjustment data of the screening scene; Generation module: Constructs scene gaze calibration monitoring domain based on parameter attribute type, and generates calibration standard items for scene gaze calibration monitoring domain based on multi-dimensional scene benchmark parameters and student gaze tracking calibration parameters; Analysis module: Generates calibration-adaptive color vision monitoring items based on real-time monitoring data of gaze trajectory and dynamic adjustment data of screening scenario. Compares and analyzes the monitoring attribute type of calibration-adaptive color vision monitoring items with the scene gaze calibration monitoring domain to obtain the target calibration monitoring domain corresponding to the calibration-adaptive color vision monitoring items. Comparison module: Compares and analyzes the calibration-adapted color vision monitoring items with the calibration standard items of the target calibration monitoring domain to obtain matching calibration items and missing calibration items; compares and analyzes the matching calibration items with the corresponding calibration standard items to obtain the calibration anomalies in the matching calibration items; Processing module: Analyzes missing calibration items to obtain the first calibration anomaly coefficient of color vision screening in the target calibration monitoring domain; analyzes calibration anomalies to obtain the second calibration anomaly coefficient of color vision screening in the target calibration monitoring domain; Output module: Obtains a comprehensive calibration anomaly coefficient based on the first and second calibration anomaly coefficients; outputs a color vision screening report based on the comprehensive calibration anomaly coefficient.

[0015] Compared with the prior art, the present invention has the following beneficial effects: This invention acquires multi-dimensional scene baseline parameters and student gaze tracking calibration parameters, combines these parameters to construct a monitoring domain and generate calibration standard items. Based on real-time gaze trajectory and scene adjustment data, it generates adapted monitoring items. This means the screening process not only conforms to the standard configuration requirements of the scene but also matches the student's actual gaze state, avoiding errors caused by fixed screening scenes or single monitoring dimensions, and allowing the screening results to better reflect the student's true color vision. Through hierarchical analysis of matching calibration items, missing calibration items, and calibration anomaly items, the first calibration anomaly coefficient and the second calibration anomaly coefficient are calculated respectively. Based on the first and second calibration anomaly coefficients, a comprehensive calibration anomaly coefficient is obtained. This allows the screening report to not only reflect the overall degree of color vision abnormality but also clearly identify whether the abnormality stems from a lack of scene configuration, improving the efficiency and scalability of color vision screening. Attached Figure Description

[0016] Figure 1 This is a schematic diagram illustrating the steps of a multimodal interactive color vision screening method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a multimodal interactive color vision screening system provided in an embodiment of the present invention. Detailed Implementation

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0018] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0019] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0020] Reference Figures 1-2 As shown.

[0021] The embodiments further illustrate the multimodal interactive color vision screening system and method proposed in this invention.

[0022] A multimodal interactive color vision screening method, comprising the following steps: Acquire multi-dimensional scene baseline parameters, student gaze tracking calibration parameters, and parameter attribute types for color vision screening; capture real-time monitoring data of the target student's gaze trajectory and dynamic adjustment data of the screening scene; Construct a scene gaze calibration monitoring domain based on parameter attribute types, and generate calibration standard items for the scene gaze calibration monitoring domain based on multi-dimensional scene benchmark parameters and student gaze tracking calibration parameters; Based on real-time monitoring data of eye trajectory and dynamic adjustment data of screening scenario, calibration and adaptation color vision monitoring items are generated. The monitoring attribute type of calibration and adaptation color vision monitoring items is compared and analyzed with the scene eye calibration monitoring domain to obtain the target calibration monitoring domain corresponding to the calibration and adaptation color vision monitoring items. The calibration-adapted color vision monitoring items are compared and analyzed with the calibration standard items of the target calibration monitoring domain to obtain the matching calibration items and the missing calibration items; the matching calibration items are compared and analyzed with the corresponding calibration standard items to obtain the calibration anomalies in the matching calibration items. The first calibration anomaly coefficient of color vision screening in the target calibration monitoring domain was obtained based on the analysis of missing calibration items; the second calibration anomaly coefficient of color vision screening in the target calibration monitoring domain was obtained based on the analysis of calibration anomalies. The comprehensive calibration anomaly coefficient is obtained based on the first calibration anomaly coefficient and the second calibration anomaly coefficient. A color vision screening report is generated based on the comprehensive calibration anomaly coefficient.

[0023] The screening status levels corresponding to the comprehensive calibration anomaly coefficient are clearly defined, with different ranges of comprehensive calibration anomaly coefficients corresponding to different color vision screening result levels. For example, a comprehensive calibration anomaly coefficient less than 1 corresponds to a normal color vision screening status; a comprehensive calibration anomaly coefficient between 1 and 3 corresponds to a mild anomaly status; a comprehensive calibration anomaly coefficient between 3 and 5 corresponds to a moderate anomaly status; and a comprehensive calibration anomaly coefficient greater than 5 corresponds to a severe anomaly status.

[0024] The report matches the specific value of the comprehensive calibration anomaly coefficient with the corresponding screening conclusion and supplements the details of the anomaly. If the comprehensive calibration anomaly coefficient is 4.6, which is in the range of 3 to 5, the corresponding screening conclusion is moderate anomaly. In this case, the report explains the source of the anomaly: for example, the anomaly in the scene reference type monitoring domain contributes significantly, specifically involving the ambient light intensity parameter deviating from the reference, the display interface adaptation parameter not meeting the standard, and the gaze calibration type monitoring domain having a gaze focusing accuracy exceeding the standard.

[0025] Based on the details of the anomalies, corresponding suggestions are provided. For example, regarding the issue of ambient light intensity deviation, it is recommended to adjust the lighting equipment in the screening scene to the reference light intensity; regarding the issue of inadequate display interface adaptation parameters, it is recommended to replace the display device with one that meets the resolution and color gamut requirements; regarding the issue of excessive gaze focus accuracy, it is recommended to guide students to keep their gaze stable during the screening process.

[0026] The report integrates screening status grading, abnormal details, and optimization suggestions to generate a complete color vision screening report. This report includes both the overall conclusions corresponding to the comprehensive calibration of abnormality coefficients and a clear definition of the specific dimensions of the abnormalities and directions for improvement, providing a reference for subsequent color vision screening adjustments or interventions.

[0027] Multi-dimensional scene benchmark parameters include color stimulus source parameters, ambient light intensity benchmark parameters, display interface adaptation parameters, and spatial layout benchmark parameters; The calibration parameters for student eye tracking include eye focus parameters, eye movement rate parameters, eye dwell time parameters, and eye deviation threshold parameters. Among them, the multi-dimensional scene benchmark parameters and student gaze tracking calibration parameters both include the corresponding benchmark update cycle; The parameter attribute types include scene baseline type and line-of-sight calibration type.

[0028] Color stimulus parameters are the core content carrier of color vision screening. For example, they specify the type of color combination used in the screening and the color contrast threshold. For instance, when screening for red-green color vision abnormalities, the color stimulus parameters are set so that the brightness difference between red and green is no less than 20%, ensuring that the stimulus can effectively distinguish between normal and abnormal color perception. Ambient light intensity benchmark parameters are the light standards for the screening environment. For example, the light intensity of the screening scene must be stable at around 500 lux to avoid excessively strong or weak light affecting color presentation and thus interfering with the accuracy of the screening results. Display interface adaptation parameters are the display specifications for the equipment used in the screening. For example, the screen resolution must be no less than 1920×1080, and the screen's color gamut must cover more than 95% of sRGB to ensure color reproduction on the display. Spatial layout benchmark parameters are the physical spatial requirements of the screening scene. For example, the distance between students and the display device must be between 50 and 70 centimeters, and the display angle of the device must be horizontal with the student's line of sight to avoid color perception errors caused by viewing angle deviations.

[0029] Eye tracking calibration parameters serve as the standard for monitoring students' gaze during screening. Eye focus parameters measure the accuracy of the student's gaze placement; for example, the deviation between the gaze placement and the target area should not exceed 5 pixels to determine if the student is accurately focusing on the screening content. Eye movement rate parameters limit the range of student gaze speed; for example, in color-changing screening, the rate should be between 10 and 20 pixels per second. Movements that are too fast or too slow may indicate abnormal color perception. Eye fixation duration parameters define the reasonable time for a student to focus on a single screening element; for example, the fixation time on a single-color pattern should be between 1 and 3 seconds. Too short a fixation means inaccurate color recognition, while too long a fixation indicates an obstacle to color discrimination. Eye deviation threshold parameters define the critical standard for the gaze leaving the screening area; for example, if the gaze deviates from the target area for more than 2 seconds, it is considered an invalid screening state, requiring readjustment of the screening process.

[0030] Both types of parameters include corresponding baseline update cycles. For example, the update cycle for multi-dimensional scenario baseline parameters is set to once every 3 months. When screening equipment is replaced or the environmental layout is adjusted, an immediate update will also be triggered. The update cycle for student eye tracking calibration parameters can be set to once per semester, and the corresponding parameter standards can be adjusted for students of different age groups.

[0031] Scene baseline types correspond to multi-dimensional scene baseline parameters, and gaze calibration types correspond to student gaze tracking calibration parameters. Through clear attribute classification, the corresponding monitoring domains are subsequently constructed based on the types, allowing different types of parameters to play their respective baseline roles in color vision screening.

[0032] Based on real-time monitoring data of eye movement trajectory and dynamic adjustment data of screening scenarios, calibrated and adapted color vision monitoring items are generated, which specifically includes the following steps: Based on parameter attribute types, scene baseline monitoring parameters and line of sight calibration monitoring parameters are extracted from real-time line of sight trajectory monitoring data and dynamic adjustment data of screening scenarios. Generate scene benchmark monitoring items based on scene benchmark monitoring parameters; Generate line-of-sight calibration monitoring items based on line-of-sight calibration monitoring parameters; The calibration and adaptation color vision monitoring items for color vision screening are composed of scene baseline monitoring items and line-of-sight calibration monitoring items.

[0033] The parameter attribute types are divided into scene baseline type and gaze calibration type, and the data sources are real-time gaze trajectory monitoring data and dynamic adjustment data of the screening scene. For example, when the parameter attribute type is scene baseline type, information such as the real-time color combination of the visual stimulus, the actual value of the current ambient light intensity, the real-time resolution and color gamut status of the display interface, and the actual distance between the student and the device in the spatial layout are extracted from the dynamic adjustment data of the screening scene. This information together constitutes the scene baseline monitoring parameters. When the parameter attribute type is gaze calibration type, information such as the deviation value between the student's gaze point and the target area, the speed of gaze movement between different elements, the duration of fixation on a single color pattern, and the duration of gaze separation from the target area are extracted from the real-time gaze trajectory monitoring data. This information constitutes the gaze calibration monitoring parameters.

[0034] Scene benchmark monitoring items are generated based on the extracted scene benchmark monitoring parameters. If the scene benchmark monitoring parameters include a current ambient light intensity of 300 lux, while the corresponding multi-dimensional scene benchmark parameters include an ambient light intensity benchmark of 500 lux, the difference between the current ambient light intensity and the benchmark light intensity is taken as a scene benchmark monitoring item. If the benchmark for the display interface adaptation parameters is a resolution of not less than 1920×1080, while the real-time extracted display interface resolution is 1366×768, the display interface resolution not meeting the benchmark is taken as another scene benchmark monitoring item.

[0035] A gaze calibration monitoring item is generated based on the gaze calibration monitoring parameters. If the gaze calibration monitoring parameters include a deviation of 8 pixels between the student's gaze point and the target area, and the corresponding gaze focusing accuracy parameter in the student gaze tracking calibration parameters is no more than 5 pixels, then the gaze focusing accuracy exceeding the benchmark range is considered as a gaze calibration monitoring item. If the benchmark for the gaze dwell time parameter is 1 to 3 seconds, and the real-time extracted gaze dwell time is 5 seconds, then the gaze dwell time exceeding the benchmark range is considered as another gaze calibration monitoring item.

[0036] The generated scene baseline monitoring items and gaze calibration monitoring items are integrated to form the calibration-adaptive color vision monitoring items for color vision screening. These monitoring items cover the actual situation in two dimensions: scene configuration and student gaze status, providing a complete monitoring basis for subsequent comparison with benchmark standards and judgment of the calibration status of color vision screening.

[0037] The monitoring attribute type of the calibration-adapted color vision monitoring item is compared and analyzed with the scene gaze calibration monitoring domain to obtain the target calibration monitoring domain corresponding to the calibration-adapted color vision monitoring item. The specific steps include: Compare the attribute types of the monitoring attribute types corresponding to the calibration and adaptation color vision monitoring items with the parameter attribute types corresponding to the scene vision calibration monitoring domain; The scene gaze calibration monitoring field with the same parameter attribute type as the monitoring attribute type is marked as the target calibration monitoring field corresponding to the calibration adaptation color vision monitoring item.

[0038] The monitoring attribute types of calibration-adapted color vision monitoring items are compared with the parameter attribute types of scene gaze calibration monitoring domains. The monitoring attribute type represents the category to which the calibration-adapted color vision monitoring item belongs. For example, if a calibration-adapted color vision monitoring item is generated based on scene baseline monitoring parameters, its monitoring attribute type belongs to the scene baseline type. The parameter attribute type of the scene gaze calibration monitoring domain, on the other hand, is the category upon which the monitoring domain was initially constructed. For instance, some scene gaze calibration monitoring domains are constructed based on parameters of the scene baseline type, and their parameter attribute type is the scene baseline type; others are constructed based on parameters of the gaze calibration type, and their parameter attribute type is the gaze calibration type.

[0039] The system filters out scene gaze calibration monitoring domains whose parameter attribute types match the monitoring attribute types and marks them as the target calibration monitoring domains corresponding to the calibration-adapted color vision monitoring items. If the monitoring attribute type of a calibration-adapted color vision monitoring item is a scene baseline type, the system finds a monitoring domain among all scene gaze calibration monitoring domains whose parameter attribute type is also a scene baseline type, and this monitoring domain is the target calibration monitoring domain corresponding to the monitoring item. Similarly, if the monitoring attribute type of a monitoring item is a gaze calibration type, the system matches scene gaze calibration monitoring domains with parameter attribute types of gaze calibration type as its target calibration monitoring domains.

[0040] Through such attribute type comparison and matching, each calibration-adapted color vision monitoring item can correspond to the target calibration monitoring domain.

[0041] The calibration-adapted color vision monitoring items are compared and analyzed with the calibration standard items of the target calibration monitoring domain to obtain the matching calibration items and the missing calibration items. The specific steps include: Obtain the benchmark update cycle and historical calibration data corresponding to the calibration standard items in the target calibration monitoring domain; The calibration standard items to be calibrated are obtained based on the benchmark update cycle and historical calibration data; The calibration-adapted color vision monitoring items are compared with the calibration standard items of the target calibration monitoring domain. Mark the calibration-adaptive color vision monitoring items that match the calibration standard items in the target calibration monitoring domain as matching calibration items; Mark the standard items to be calibrated in the target calibration monitoring domain that do not match any of the calibration-adapted color vision monitoring items as missing calibration items.

[0042] Obtain the baseline update cycle and historical calibration data for each calibration standard item within the target calibration monitoring domain. The baseline update cycle is the time interval at which the calibration standard item needs to be updated, for example, the baseline update cycle for a certain calibration standard item is 3 months; historical calibration data is the past calibration records for the standard item, such as the parameter values ​​and calibration time information of the past 3 calibrations.

[0043] The calibration standard items to be calibrated are determined based on the benchmark update cycle and historical calibration data. If the benchmark update cycle of a calibration standard item is 3 months, and its most recent calibration time is more than 3 months ago, it is determined to be a calibration standard item based on the validity record of the standard item in the historical calibration data; if the most recent calibration time of a standard item is within the benchmark update cycle, and the historical calibration data shows that its parameters are stable, it will not be included in the calibration standard items.

[0044] The calibration-adapted color vision monitoring items are compared item by item with the calibration standard items in the target calibration monitoring domain. For example, if the calibration-adapted color vision monitoring items include the monitoring of the current ambient light intensity, the standard item corresponding to the ambient light intensity reference parameter is found in the calibration standard items of the target calibration monitoring domain, and the parameter contents of the two are matched and verified.

[0045] A calibration-adapted color vision monitoring item that matches the content of the calibration standard item in the target calibration monitoring domain is defined as a matched calibration item. If the current ambient light intensity parameter in the calibration-adapted color vision monitoring item is consistent with the content of the ambient light intensity reference parameter standard item in the target monitoring domain, the monitoring item is marked as a matched calibration item.

[0046] In the target calibration monitoring domain, if a standard item cannot be matched with any of the calibration-adapted color vision monitoring items, that standard item will be marked as a missing calibration item. For example, if the target monitoring domain includes display interface adaptation parameters as a standard item to be calibrated, but the calibration-adapted color vision monitoring items do not contain the corresponding monitoring content, then that display interface adaptation parameter standard item will be determined as a missing calibration item.

[0047] The analysis of missing calibration items yields the first calibration anomaly coefficient for color vision screening in the target calibration monitoring domain, specifically including the following steps: Obtain the number of missing calibration items and the number of parameters to be calibrated in the target calibration monitoring domain; The missing percentage of the missing calibration items in the target calibration monitoring domain is obtained based on the number of missing items and the number of items to be calibrated. The missing percentage coefficient of the missing calibration items is obtained based on the missing percentage value. The missing calibration period value of the missing calibration item is obtained based on the benchmark update cycle of the missing calibration item and historical calibration data, and the missing calibration period coefficient of the missing calibration item is obtained based on the missing calibration period value. The missing anomaly coefficient of the missing calibration item is obtained based on the missing percentage coefficient and the missing period coefficient. The missing anomaly coefficients of the missing calibration items in the target calibration monitoring domain are all accumulated to obtain the first calibration anomaly coefficient of color vision screening in the target calibration monitoring domain.

[0048] The number of missing calibration items is the number of standard items to be calibrated in the target calibration monitoring domain that do not match any calibration-adaptive color vision monitoring item; the number of calibrated parameter items in the target calibration monitoring domain is the total number of standard items that need to be calibrated in that domain. For example, if a target calibration monitoring domain has 10 items to be calibrated, and 3 of them do not match any monitoring item, then the number of missing items is 3.

[0049] The missing percentage value is calculated based on the number of missing items and the number of items to be calibrated, resulting in the missing percentage coefficient. Missing percentage value = Number of missing items ÷ Number of items to be calibrated. For example, when the number of missing items is 3 and the number of items to be calibrated is 10, the missing percentage value is 3 ÷ 10 = 0.3; the missing percentage coefficient is usually taken directly from the missing percentage value, so the missing percentage coefficient is 0.3.

[0050] Based on the baseline update cycle of the missing calibration item and historical calibration data, calculate the missing calibration cycle value, and then convert it into a missing cycle coefficient. Missing calibration cycle value = (current time - most recent valid calibration time) ÷ baseline update cycle. For example, if the baseline update cycle for a missing calibration item is 3 months, and the most recent valid calibration time is 6 months ago, then the missing calibration cycle value is 6 ÷ 3 = 2; the missing cycle coefficient is generally taken as the value of this cycle, i.e., in this case, the missing cycle coefficient is 2.

[0051] The missing anomaly coefficient for a single missing calibration item is calculated by combining the missing percentage coefficient and the missing period coefficient. Missing anomaly coefficient = Missing percentage coefficient × Missing period coefficient. When the missing percentage coefficient is 0.3 and the missing period coefficient is 2, the missing anomaly coefficient for this calibration item is 0.3 × 2 = 0.6.

[0052] The first calibration anomaly coefficient for color vision screening in the target calibration monitoring domain is obtained by summing the missing anomaly coefficients of all missing calibration items in the target calibration monitoring domain. For example, if there are 3 missing calibration items in this domain, with corresponding missing anomaly coefficients of 0.6, 0.4, and 0.5, then the first calibration anomaly coefficient is 0.6 + 0.4 + 0.5 = 1.5.

[0053] The second calibration anomaly coefficient for color vision screening in the target calibration monitoring domain is obtained based on calibration anomaly analysis, specifically including the following steps: The parameter anomaly coefficient is obtained by comparing the calibration monitoring parameters of the matching calibration item with the standard parameters of the corresponding calibration standard item. The parameter anomaly coefficients corresponding to the calibration anomalies in the target calibration monitoring domain are summed to obtain the second calibration anomaly coefficient of color vision screening in the target calibration monitoring domain.

[0054] The parameter anomaly coefficient is obtained by comparing the calibration monitoring parameters of the matched calibration item with the standard parameters of the corresponding calibration standard item. The calibration monitoring parameters are the actual values ​​collected by the matched calibration item, while the standard parameters are the preset reference values ​​of the calibration standard item. The calculation of the parameter anomaly coefficient needs to be determined in conjunction with the parameter type. For example, for numerical parameters such as ambient light intensity, the parameter anomaly coefficient = |calibration monitoring parameter - standard parameter| ÷ standard parameter. If the standard parameter corresponding to a matched calibration item is ambient light intensity of 500 lux, while its calibration monitoring parameter is 300 lux, then the parameter anomaly coefficient of the matched item is |300-500| ÷ 500 = 0.4; if it is a threshold-type parameter such as line-of-sight focusing accuracy, the standard parameter is no more than 5 pixels, while the calibration monitoring parameter is 8 pixels, then the parameter anomaly coefficient is (8-5) ÷ 5 = 0.6.

[0055] Calibration anomalies are selected from the matched calibration items; these are those where the calibration monitoring parameters do not conform to the standard parameters. The parameter anomaly coefficients corresponding to all calibration anomalies within the target calibration monitoring domain are summed to obtain the second calibration anomaly coefficient for color vision screening within that target calibration monitoring domain. For example, if there are two calibration anomalies within a target calibration monitoring domain, with the first anomaly having a parameter anomaly coefficient of 0.4 and the second anomaly having a parameter anomaly coefficient of 0.6, then the second calibration anomaly coefficient is 0.4 + 0.6 = 1.0.

[0056] The parameter anomaly coefficient is obtained by comparing the calibration monitoring parameters of the matched calibration item with the standard parameters of the corresponding calibration standard item. This process includes the following steps: If the calibration monitoring parameters of the matched calibration item meet the standard parameters, then the matched calibration item is judged to be a normal calibration item; If the calibration monitoring parameters of the matched calibration item do not meet the standard parameters, the matched calibration item is judged to be a calibration anomaly. Based on the calibration monitoring parameters of the calibration anomaly and the standard parameters of the corresponding calibration standard, the parameter anomaly coefficient corresponding to the calibration anomaly is obtained.

[0057] When the calibration monitoring parameters of a matched calibration item meet the standard parameters of the corresponding calibration standard item, the matched calibration item is determined to be a normal calibration item; conversely, when the calibration monitoring parameters of a matched calibration item do not meet the standard parameters, the matched calibration item is determined to be a abnormal calibration item. For example, if the standard parameter corresponding to a matched calibration item is an ambient light intensity of 500 lux, and its calibration monitoring parameter is 500 lux, meeting the standard parameter requirement, then the matched item is a normal calibration item; if its calibration monitoring parameter is 300 lux, which does not meet the standard parameter of 500 lux, then the matched item is determined to be a abnormal calibration item.

[0058] Calculate the parameter anomaly coefficient for each calibration anomaly. Parameter anomaly coefficient = |calibration monitoring parameter - standard parameter| ÷ standard parameter. If the calibration monitoring parameter for an anomaly is 300 lux and the corresponding standard parameter is 500 lux, then the parameter anomaly coefficient for this anomaly is |300-500| ÷ 500 = 0.4. If it is a threshold-type parameter such as line-of-sight focusing accuracy, the standard parameter is no more than 5 pixels, while the calibration monitoring parameter is 8 pixels, then the parameter anomaly coefficient is (8-5) ÷ 5 = 0.6.

[0059] The comprehensive calibration anomaly coefficient is obtained based on the first calibration anomaly coefficient and the second calibration anomaly coefficient, specifically including the following steps: The calibration anomaly coefficient of the color vision screening in the target calibration monitoring domain is obtained based on the first calibration anomaly coefficient and the second calibration anomaly coefficient. Set the monitoring domain weights corresponding to the target calibration monitoring domain according to the parameter attribute types corresponding to the target calibration monitoring domain; The monitoring domain calibration anomaly coefficient of the target calibration monitoring domain is obtained by assigning weights to the monitoring domain weights. The anomaly coefficients of the monitoring fields of each scene's gaze calibration monitoring field are summed to obtain the comprehensive calibration anomaly coefficient for color vision screening.

[0060] By combining the first and second calibration anomaly coefficients, the monitoring domain calibration anomaly coefficient for color vision screening in the target calibration monitoring domain is obtained. Monitoring domain calibration anomaly coefficient = first calibration anomaly coefficient + second calibration anomaly coefficient. For example, if the first calibration anomaly coefficient for a target calibration monitoring domain is 1.5 and the second calibration anomaly coefficient is 1.0, then the monitoring domain calibration anomaly coefficient for that domain is 1.5 + 1.0 = 2.5.

[0061] The monitoring domain weight is set according to the parameter attribute type corresponding to the target calibration monitoring domain. The parameter attribute type is divided into scene reference type and gaze calibration type. Different types of monitoring domains have different degrees of influence on the color vision screening results, so the weights will be different. If the parameter attribute type of the target calibration monitoring domain is scene reference type, it directly affects the effectiveness of the color vision stimulus source, and its monitoring domain weight is set to 1.2; if the parameter attribute type is gaze calibration type, it reflects the student's attention status, and the monitoring domain weight is set to 0.8.

[0062] The anomaly coefficient of the target calibration monitoring domain is calculated by assigning weights to the monitoring domain weights. The anomaly coefficient is calculated as follows: Monitoring Domain Anomaly Coefficient = Monitoring Domain Calibration Anomaly Coefficient × Monitoring Domain Weight. For example, if the monitoring domain calibration anomaly coefficient of a certain scenario baseline type monitoring domain is 2.5 and the weight is 1.2, then its monitoring domain anomaly coefficient is 2.5 × 1.2 = 3.0; if the monitoring domain calibration anomaly coefficient of a certain line-of-sight calibration type monitoring domain is 2.0 and the weight is 0.8, then its monitoring domain anomaly coefficient is 2.0 × 0.8 = 1.6.

[0063] The overall calibration anomaly coefficient for color vision screening is obtained by summing the anomaly coefficients of all scene gaze calibration monitoring domains. For example, if the entire screening process includes two monitoring domains: scene baseline type and gaze calibration type, with corresponding monitoring domain anomaly coefficients of 3.0 and 1.6 respectively, then the overall calibration anomaly coefficient is 3.0 + 1.6 = 4.6.

[0064] A multimodal interactive color vision screening system, comprising: Acquisition Module: Acquires multi-dimensional scene baseline parameters for color vision screening, student gaze tracking calibration parameters, and parameter attribute types; captures real-time monitoring data of the target student's gaze trajectory and dynamic adjustment data of the screening scene; Generation module: Constructs scene gaze calibration monitoring domain based on parameter attribute type, and generates calibration standard items for scene gaze calibration monitoring domain based on multi-dimensional scene benchmark parameters and student gaze tracking calibration parameters; Analysis module: Generates calibration-adaptive color vision monitoring items based on real-time monitoring data of gaze trajectory and dynamic adjustment data of screening scenario. Compares and analyzes the monitoring attribute type of calibration-adaptive color vision monitoring items with the scene gaze calibration monitoring domain to obtain the target calibration monitoring domain corresponding to the calibration-adaptive color vision monitoring items. Comparison module: Compares and analyzes the calibration-adapted color vision monitoring items with the calibration standard items of the target calibration monitoring domain to obtain matching calibration items and missing calibration items; compares and analyzes the matching calibration items with the corresponding calibration standard items to obtain the calibration anomalies in the matching calibration items; Processing module: Analyzes missing calibration items to obtain the first calibration anomaly coefficient of color vision screening in the target calibration monitoring domain; analyzes calibration anomalies to obtain the second calibration anomaly coefficient of color vision screening in the target calibration monitoring domain; Output module: Obtains a comprehensive calibration anomaly coefficient based on the first and second calibration anomaly coefficients; outputs a color vision screening report based on the comprehensive calibration anomaly coefficient.

[0065] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0066] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0067] 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 multi-modal interactive color vision screening method, characterized in that, The method comprises the following steps: Obtaining multi-dimensional scene reference parameters of color vision screening, student visual line tracking calibration parameters and parameter attribute types; capturing visual line trajectory real-time monitoring data and screening scene dynamic adjustment data of a target student; According to the parameter attribute types, a scene visual line calibration monitoring domain is constructed, and calibration standard items of the scene visual line calibration monitoring domain are generated according to the multi-dimensional scene reference parameters and the student visual line tracking calibration parameters; According to the visual line trajectory real-time monitoring data and the screening scene dynamic adjustment data, calibration adaptive color vision monitoring items are generated, the monitoring attribute types of the calibration adaptive color vision monitoring items are compared and analyzed with the scene visual line calibration monitoring domain, and a target calibration monitoring domain corresponding to the calibration adaptive color vision monitoring items is obtained; The calibration adaptive color vision monitoring items are compared and analyzed with the calibration standard items of the target calibration monitoring domain, matching calibration items and missing calibration items are obtained, the matching calibration items are compared and analyzed with the calibration standard items corresponding to the matching calibration items, and calibration abnormal items in the matching calibration items are obtained; A first calibration abnormal coefficient of the color vision screening in the target calibration monitoring domain is obtained according to the missing calibration items, and a second calibration abnormal coefficient of the color vision screening in the target calibration monitoring domain is obtained according to the calibration abnormal items; A comprehensive calibration abnormal coefficient is obtained according to the first calibration abnormal coefficient and the second calibration abnormal coefficient, and a color vision screening report is output according to the comprehensive calibration abnormal coefficient.

2. The multi-modal interactive color vision screening method of claim 1, wherein, The multi-dimensional scene reference parameters comprise color vision stimulus source parameters, environmental light intensity reference parameters, display interface adaptive parameters and spatial layout reference parameters; The student visual line tracking calibration parameters comprise visual line focusing parameters, visual line moving rate parameters, visual line staying time length parameters and visual line deviation threshold parameters; The multi-dimensional scene reference parameters and the student visual line tracking calibration parameters both comprise corresponding reference update periods. The parameter attribute types comprise scene reference types and visual line calibration types.

3. The multi-modal interactive color vision screening method of claim 2, wherein, The calibration adaptive color vision monitoring items are generated according to the visual line trajectory real-time monitoring data and the screening scene dynamic adjustment data, and specifically comprise the following steps: Scene reference monitoring parameters and visual line calibration monitoring parameters are extracted from the visual line trajectory real-time monitoring data and the screening scene dynamic adjustment data based on the parameter attribute types; Scene reference monitoring items are generated according to the scene reference monitoring parameters; Visual line calibration monitoring items are generated according to the visual line calibration monitoring parameters; The calibration adaptive color vision monitoring items of the color vision screening are constituted by the scene reference monitoring items and the visual line calibration monitoring items.

4. The multi-modal interactive color vision screening method of claim 3, wherein, The monitoring attribute types of the calibration adaptive color vision monitoring items are compared and analyzed with the scene visual line calibration monitoring domain, and a target calibration monitoring domain corresponding to the calibration adaptive color vision monitoring items is obtained, specifically comprising the following steps: Attribute type comparison is performed between the monitoring attribute types corresponding to the calibration adaptive color vision monitoring items and the parameter attribute types corresponding to the scene visual line calibration monitoring domain; The scene visual line calibration monitoring domains with the same parameter attribute types and monitoring attribute types are marked as the target calibration monitoring domains corresponding to the calibration adaptive color vision monitoring items.

5. The multi-modal interactive color vision screening method of claim 4, wherein, The calibration adaptive color vision monitoring items are compared and analyzed with the calibration standard items of the target calibration monitoring domain, and matching calibration items and missing calibration items are obtained, specifically comprising the following steps: Obtaining a reference update period and historical calibration data corresponding to the calibration standard item in the target calibration monitoring domain; Obtaining a calibration standard item to be calibrated from the reference update period and the historical calibration data; Comparing the calibration adaptive color vision monitoring item with the calibration standard item in the target calibration monitoring domain; Labeling the calibration adaptive color vision monitoring item matching the calibration standard item in the target calibration monitoring domain as a matching calibration item; Labeling the calibration standard item to be calibrated in the target calibration monitoring domain as a missing calibration item.

6. The multi-modal interactive color vision screening method of claim 5, wherein, According to the missing calibration item, a first calibration abnormality coefficient of the color vision screening in the target calibration monitoring domain is obtained, which specifically includes the following steps: Obtaining the number of missing items and the number of calibration parameter items to be calibrated in the target calibration monitoring domain; Obtaining a missing proportion value of the missing calibration item in the target calibration monitoring domain according to the number of missing items and the number of calibration parameter items to be calibrated, and obtaining a missing proportion coefficient of the missing calibration item according to the missing proportion value; Obtaining a missing calibration period value of the missing calibration item according to the reference update period and the historical calibration data of the missing calibration item, and obtaining a missing period coefficient of the missing calibration item according to the missing calibration period value; Obtaining a missing abnormality coefficient of the missing calibration item according to the missing proportion coefficient and the missing period coefficient, and accumulating the missing abnormality coefficients of the missing calibration items in the target calibration monitoring domain to obtain the first calibration abnormality coefficient of the color vision screening in the target calibration monitoring domain.

7. The multi-modal interactive color vision screening method of claim 6, wherein, According to the calibration abnormality item, a second calibration abnormality coefficient of the color vision screening in the target calibration monitoring domain is obtained, which specifically includes the following steps: Comparing the calibration monitoring parameter of the matching calibration item with the standard parameter corresponding to the matching calibration item to obtain a parameter abnormality coefficient; Accumulating the parameter abnormality coefficients corresponding to the calibration abnormality items in the target calibration monitoring domain to obtain the second calibration abnormality coefficient of the color vision screening in the target calibration monitoring domain.

8. The multi-modal interactive color vision screening method of claim 7, wherein, Comparing the calibration monitoring parameter of the matching calibration item with the standard parameter corresponding to the matching calibration item to obtain a parameter abnormality coefficient, which specifically includes the following steps: If the calibration monitoring parameter of the matching calibration item meets the standard parameter, the matching calibration item is determined to be a calibration normal item; If the calibration monitoring parameter of the matching calibration item does not meet the standard parameter, the matching calibration item is determined to be a calibration abnormality item; Obtaining the parameter abnormality coefficient corresponding to the calibration abnormality item according to the calibration monitoring parameter of the calibration abnormality item and the standard parameter corresponding to the calibration abnormality item.

9. The multi-modal interactive color vision screening method of claim 8, wherein, Obtaining a comprehensive calibration abnormality coefficient according to the first calibration abnormality coefficient and the second calibration abnormality coefficient, which specifically includes the following steps: Obtaining a monitoring domain calibration abnormality coefficient of the color vision screening in the target calibration monitoring domain according to the first calibration abnormality coefficient and the second calibration abnormality coefficient; Setting a monitoring domain weight value corresponding to the target calibration monitoring domain according to the parameter attribute type corresponding to the target calibration monitoring domain; Weighting the monitoring domain calibration abnormality coefficient of the target calibration monitoring domain based on the monitoring domain weight value to obtain a monitoring domain abnormality coefficient corresponding to the target calibration monitoring domain; Accumulating the monitoring domain abnormality coefficients of each scene line-of-sight calibration monitoring domain to obtain a comprehensive calibration abnormality coefficient of the color vision screening.

10. A multi-modal interactive color vision screening system for use in a multi-modal interactive color vision screening method according to any one of claims 1 to 9, characterized in that, It includes: The acquisition module acquires multi-dimensional scene reference parameters of color vision screening, student visual line tracking calibration parameters and parameter attribute types; Real-time monitoring data of visual line trajectory of the target student and dynamic adjustment data of the screening scene are captured; The generation module constructs a scene visual line calibration monitoring domain according to the parameter attribute types, and generates calibration standard items of the scene visual line calibration monitoring domain according to the multi-dimensional scene reference parameters and the student visual line tracking calibration parameters; The analysis module generates calibration adaptive color vision monitoring items according to the real-time monitoring data of the visual line trajectory and the dynamic adjustment data of the screening scene, compares and analyzes monitoring attribute types of the calibration adaptive color vision monitoring items with the scene visual line calibration monitoring domain, and obtains target calibration monitoring domains corresponding to the calibration adaptive color vision monitoring items; The comparison module compares and analyzes the calibration adaptive color vision monitoring items with the calibration standard items of the target calibration monitoring domains, and obtains matching calibration items and missing calibration items; The matching calibration items are compared and analyzed with the calibration standard items corresponding to the matching calibration items, and calibration abnormal items in the matching calibration items are obtained; The processing module analyzes the first calibration abnormal coefficient of the color vision screening in the target calibration monitoring domain according to the missing calibration items, and analyzes the second calibration abnormal coefficient of the color vision screening in the target calibration monitoring domain according to the calibration abnormal items; The output module obtains a comprehensive calibration abnormal coefficient according to the first calibration abnormal coefficient and the second calibration abnormal coefficient, and outputs a color vision screening report according to the comprehensive calibration abnormal coefficient.

Images