Reading ability assessment method and device, equipment and storage medium

Abstract

This application discloses a reading ability assessment method, apparatus, device, and storage medium. It acquires raw eye-tracking data of the test subject during the reading of assessment materials, segments the raw eye-tracking data into a set of fixations, and obtains eye-tracking features corresponding to each reading ability indicator based on the regions of interest and the set of fixations on the assessment materials. This allows for the determination of the test subject's level under the corresponding reading ability indicator, and further, based on the level under each reading ability indicator, the test subject's overall reading ability level is determined. This application determines the test subject's overall reading ability level based on the test subject's actual reading process, resulting in higher accuracy. This application can measure the test subject's reading ability level from multiple different dimensions, such as reading comprehension, reading attention, and reading memory, leading to a more interpretable and accurate overall reading ability level that better reflects the test subject's actual reading ability.

Description

Technical Field

[0001] This application relates to the field of smart education technology, and more specifically, to a reading ability assessment method, device, equipment, and storage medium. Background Technology

[0002] Reading ability is the foundation of learning ability. However, early signs of reading difficulties in children are often easily labeled by parents and teachers as a lack of effort or concentration, thus missing the optimal time for detection and correction, and harming the child's physical and mental development. Therefore, it is necessary to design a reading ability assessment program to accurately measure the reading ability level of the test taker.

[0003] In recent years, significant progress has been made in screening for dyslexia from a neuropsychological perspective. Current screening methods primarily rely on diverse paper-and-pencil / electronic psychological questionnaires. These methods largely depend on statistical analysis of questionnaire results, lacking a more granular depiction of the reading process, thus limiting screening accuracy. Summary of the Invention

[0004] In view of the above problems, this application is made to provide a reading ability assessment method, device, equipment, and storage medium to achieve a more accurate assessment of the true reading ability level of test subjects. The specific solution is as follows:

[0005] Firstly, it provides a method for assessing reading ability, including:

[0006] Obtain raw eye movement data of the test subject during the process of reading the specified assessment materials;

[0007] Based on the original eye movement data, fixation points are segmented to obtain a set of fixation points;

[0008] Based on the regions of interest and the set of fixations defined on the specified assessment material, eye movement features corresponding to each pre-configured reading ability indicator are obtained. The reading ability indicators include at least one of reading comprehension, reading attention, and reading memory.

[0009] Based on the eye movement characteristics corresponding to each reading ability indicator, the test subject's level under the corresponding reading ability indicator is determined, and based on the test subject's level under each reading ability indicator, the test subject's overall reading ability level is determined.

[0010] Secondly, a reading ability assessment device is provided, comprising:

[0011] The raw eye movement data acquisition unit is used to acquire raw eye movement data of the test subject during the process of reading the specified assessment materials;

[0012] A gaze point segmentation unit is used to segment gaze points based on the original eye movement data to obtain a set of gaze points;

[0013] An eye movement feature acquisition unit is used to acquire eye movement features corresponding to each pre-configured reading ability indicator based on the region of interest divided on the specified assessment material and the set of fixation points. The reading ability indicators include at least one of reading comprehension, reading attention, and reading memory.

[0014] The indicator level determination unit is used to determine the level of the test subject under the corresponding reading ability indicator based on the eye movement characteristics corresponding to each reading ability indicator;

[0015] The comprehensive level determination unit is used to determine the comprehensive reading ability level of the test subject based on the test subject's level under each reading ability indicator.

[0016] Thirdly, a reading ability assessment device is provided, including: a memory and a processor;

[0017] The memory is used to store programs;

[0018] The processor is used to execute the program to implement the various steps of the aforementioned reading ability assessment method.

[0019] Fourthly, a storage medium is provided on which a computer program is stored, which, when executed by a processor, implements the various steps of the aforementioned reading ability assessment method.

[0020] By employing the above technical solution, this application provides a reading ability assessment method based on eye-tracking data. This application specifies materials for reading ability assessment, requires test subjects to read these materials, and acquires raw eye-tracking data during the reading process. By segmenting the raw eye-tracking data into fixation points, a set of fixation points is obtained. Then, based on the regions of interest (ROIs) defined on the specified assessment material and the set of fixation points, eye-tracking features corresponding to pre-configured reading ability indicators are obtained. This determines the test subject's level under the corresponding reading ability indicator, and further, based on the level under each reading ability indicator, the test subject's overall reading ability level is determined. Clearly, this application starts from the test subject's actual reading process, captures the test subject's eye-tracking data, and then extracts eye-tracking features that can measure reading ability from different indicator dimensions based on the fixation points in the eye-tracking data and the ROIs on the material. This determines the test subject's level under each reading ability indicator and ultimately determines the test subject's overall reading ability level, resulting in higher accuracy.

[0021] Meanwhile, this application constructs an evaluation index system for the reading ability level of test subjects, which can measure the reading ability level of test subjects from three different indicators: reading comprehension, reading attention, and reading memory. The level of the corresponding reading ability indicator is determined based on the eye movement features extracted under each reading ability indicator. Finally, the comprehensive reading ability level is determined based on the level of each reading ability indicator. The resulting comprehensive reading ability level is more interpretable and more consistent with the actual reading ability of the test subjects. Attached Figure Description

[0022] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0023] Figure 1 A flowchart illustrating the reading ability assessment method provided in this application embodiment;

[0024] Figure 2a An example is provided: a schematic diagram of the gaze point trajectory on a passage reading material;

[0025] Figure 2b An example is provided, illustrating the gaze trajectory on a digitally named material.

[0026] Figure 3 This example illustrates a schematic diagram of a reading ability level assessment index system;

[0027] Figure 4 An example is provided, illustrating a geometric relationship for calculating the instantaneous angular velocity of the current eye-tracking data point;

[0028] Figure 5 This example illustrates the display effect of a reading ability assessment interface.

[0029] Figure 6 This example illustrates another way to display a reading ability assessment interface.

[0030] Figure 7 This is a schematic diagram of the structure of a reading ability assessment device provided in an embodiment of this application;

[0031] Figure 8 This is a schematic diagram of the structure of the reading ability assessment device provided in an embodiment of this application. Detailed Implementation

[0032] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0033] This application provides a multidimensional reading ability assessment scheme based on eye-tracking data, which can be applied to solve various types of reading ability assessment tasks, such as assessing the reading ability of school-aged children and assessing the reading ability of test subjects in reading disorder screening tasks.

[0034] The proposed solution can be implemented based on a terminal with data processing capabilities, such as a mobile phone, computer, server, or cloud platform.

[0035] Next, combined Figure 1 The reading ability assessment method described in this application may include the following steps:

[0036] Step S100: Obtain the raw eye movement data of the subject during the process of reading the specified assessment material.

[0037] In this embodiment, matching assessment materials can be selected based on the reading ability assessment task for the test taker to read. The assessment materials can be one or more.

[0038] For example, in assessing the reading ability of school-aged children (6-12 years old), two types of materials can be selected as assessment materials: passage reading and rapid number naming. Passage reading materials include... Figure 2a As shown, the reading materials include content such as: "Different ships use different power sources to move. In ancient times, cargo ships usually relied on sails to use wind power to move forward... Now, some ships have better power sources."; materials on rapid number naming include... Figure 2b As shown, the reading material contains a string of Arabic numerals, such as "5 4 2 6...9 6 2 4". Through both passage reading and rapid number naming, the reading materials can assess school-aged children's ability to process text and common numbers.

[0039] In this embodiment, an eye tracker can be used as the eye movement data acquisition device. Generally, specified assessment materials are displayed on the screen, and an eye tracker is placed below the screen to collect raw eye movement data of the test subject as they read the specified assessment materials on the screen.

[0040] Before using the eye-tracking data acquisition device, the subject's posture and eyes can be calibrated first. After calibration, the subject is prompted to start reading the designated assessment material displayed on the screen, and the subject's raw eye-tracking data is obtained through the eye-tracking data acquisition device.

[0041] The eye-tracking data acquisition device collects eye-tracking data points at a set frequency. In other words, the raw eye-tracking data can be regarded as a series of eye-tracking data points. Each eye-tracking data point represents the position point on the screen that the subject's eyes are focused on, and each eye-tracking data point carries a time tag.

[0042] Step S110: Perform fixation point segmentation based on the original eye movement data to obtain a fixation point set.

[0043] Specifically, human visual perception of objects and scenes is accomplished through a series of fixations and saccades. A saccade is a rapid eye movement that moves the foveal field of vision from one point to another, while fixation involves holding the foveal field of vision on a target for a certain duration to obtain sufficient visual image details. Because the eyeballs move extremely quickly during saccades, almost no effective visual information is obtained during this time; most visual information is acquired through fixation.

[0044] To better measure the reading ability of the test subjects, this case focuses more on fixation behavior. Therefore, it is necessary to perform fixation point segmentation on the raw eye movement data, that is, to divide each eye movement data point in the raw eye movement data into two types of points: fixation points and saccade points, so that the set of fixation points is composed of the segmented fixation points.

[0045] Step S120: Based on the region of interest divided on the specified assessment material and the set of fixation points, obtain the eye movement features corresponding to each pre-configured reading ability indicator.

[0046] Specifically, after obtaining the specified evaluation materials, the materials can first be divided into regions of interest (AOIs). Different division rules and granularities can be used when dividing the AOIs; for example, characters, phrases, or sentences can be used as basic division units, with each basic unit constituting a region of interest.

[0047] For the example in this embodiment Figure 2a For reading materials, this application provides a method for dividing the text: using punctuation marks and the end of a line of text as dividing points to define the regions of interest. Figure 2a The example reading material can be divided into 11 regions of interest.

[0048] For the example in this embodiment Figure 2bFor reading materials on rapid naming of numbers, this application provides a method for partitioning the data: x digits form a region of interest, where x can take the value 5. Figure 2b The example of rapid number naming reading material can be divided into 10 regions of interest.

[0049] It should be noted that the above process of dividing the evaluation materials into regions of interest generally refers to dividing the content areas of the evaluation materials into regions of interest. Areas in the evaluation materials other than the regions of interest generally do not contain valid content and can be considered as non-regions of interest, and are marked as other areas.

[0050] Each fixation point in the set possesses both positional and temporal order, and can be assigned to different regions of interest or other regions based on its position. Building upon this, this step can obtain eye movement characteristics corresponding to each reading ability indicator based on the regions of interest in the assessment material and the positional and temporal attributes of each fixation point in the set.

[0051] In this embodiment, reading ability indicators can be designed from multiple different dimensions. For example, the pre-configured reading ability indicators include, but are not limited to, at least one of reading comprehension, reading attention, and reading memory.

[0052] This application can pre-design eye movement features for measuring each reading ability indicator, and then extract the eye movement features for each reading ability indicator in this step based on the region of interest and the set of fixations. It should be noted that the eye movement features for each reading ability indicator may include one or more.

[0053] Step S130: Based on the eye movement characteristics corresponding to each reading ability indicator, determine the test subject's level under the corresponding reading ability indicator.

[0054] Specifically, after obtaining the eye movement characteristics corresponding to each reading ability indicator in the previous step, the test subject's level under the corresponding reading ability indicator can be determined based on the obtained eye movement characteristics. The test subject's level under the corresponding reading ability indicator can be represented by a score or rating.

[0055] For example, after obtaining the eye movement characteristics corresponding to the reading comprehension index, the reading comprehension level of the test subject can be determined based on the eye movement characteristics.

[0056] Step S140: Determine the overall reading ability level of the test subject based on the test subject's level under each reading ability indicator.

[0057] Specifically, when this application selects multiple reading ability indicators of different dimensions, the test subject's level under multiple different reading ability indicators can be obtained through the previous step, and then the test subject's comprehensive reading ability level can be determined by comprehensively considering the level under each different reading ability indicator.

[0058] The overall reading ability level of the test takers can be expressed by means of scores or ratings.

[0059] The reading ability assessment method provided in this application starts from the actual reading process of the test subject, captures the test subject's eye movement data, and then extracts eye movement features that can measure reading ability from different indicator dimensions based on the fixation point in the eye movement data and the region of interest in the material, so as to determine the test subject's level under each reading ability indicator, and finally determines the test subject's comprehensive reading ability level. The accuracy of this result is higher.

[0060] Meanwhile, this application constructs an evaluation index system for the reading ability level of test subjects, which can measure the reading ability level of test subjects from three different indicators: reading comprehension, reading attention, and reading memory. The level of the corresponding reading ability indicator is determined based on the eye movement features extracted under each reading ability indicator. Finally, the comprehensive reading ability level is determined based on the level of each reading ability indicator. The resulting comprehensive reading ability level is more interpretable and more consistent with the actual reading ability of the test subjects.

[0061] In some embodiments of this application, a specific implementation of a reading ability assessment method is provided. In this embodiment, the reading ability indicators include three dimensions: reading comprehension, reading attention, and reading memory.

[0062] The eye movement features extracted in step S120 above, corresponding to each reading ability indicator, may specifically include:

[0063] The first eye movement feature corresponds to the reading comprehension indicator, the second eye movement feature corresponds to the reading attention indicator, and the third eye movement feature corresponds to the reading memory indicator.

[0064] Based on this, the implementation process of the aforementioned steps S130 and S140 may specifically include:

[0065] Based on the first eye-tracking feature, the subject's reading comprehension level is determined; based on the second eye-tracking feature, the subject's reading attention level is determined; and based on the third eye-tracking feature, the subject's reading memory level is determined. Furthermore, based on the subject's reading comprehension level, reading attention level, and reading memory level, the subject's overall reading ability level is determined.

[0066] In this embodiment, three different reading ability indicators were designed based on three dimensions: reading comprehension, attention, and memory. The level of the test subject was evaluated from the perspective of each of the three indicators, and then the overall reading ability level of the test subject was obtained by combining the levels of the three indicators. The evaluation results are more accurate and more interpretable.

[0067] Combination Figure 3 As shown in the figure, this embodiment introduces an optional index system for assessing the reading ability level of test subjects.

[0068] In this embodiment, three reading ability indicators are used as examples: reading comprehension, reading attention, and reading memory.

[0069] Each reading ability indicator can include multiple types of eye movement features. These multiple types of eye movement features can be further divided into time-dimension eye movement features and spatial-dimension eye movement features, defined as a time feature set and a spatial feature set, respectively. Each feature set contains at least one type of eye movement feature.

[0070] This embodiment can pre-construct training data. Specifically, several typical normal test subjects and several high-risk test subjects with reading difficulties are selected. These selected test subjects are processed according to the evaluation method of the aforementioned embodiment until the eye movement characteristics of each test subject under each reading ability index are obtained. Each type of eye movement characteristic is used as a training sample, and domain experts are used to label the comprehensive reading ability level of the test subjects to obtain training data.

[0071] Based on this, exploratory factor analysis can be used to analyze the training data to obtain... Figure 3 The weights of the reading ability level assessment index system in the example are obtained from the bottom up, that is, the weights corresponding to each type of eye movement feature and the weights corresponding to each reading ability index.

[0072] It should be noted that, Figure 3 The lowest level is illustrated using a feature set as an example. Specifically, when determining the weights, the weights corresponding to each type of eye movement feature in the feature set can be determined separately. That is, the weights corresponding to different types of eye movement features in the same feature set can be determined separately.

[0073] Based on the pre-configured weights, step S130, which involves determining the test subject's level under each reading ability indicator based on the eye movement characteristics corresponding to that indicator, may include:

[0074] By using pre-configured weights corresponding to each type of eye movement feature, the eye movement features corresponding to each reading ability indicator are linearly weighted to obtain the score of each reading ability indicator, which is used as the test subject's level under the corresponding reading ability indicator.

[0075] The aforementioned step S140, the process of determining the test subject's overall reading ability level based on the test subject's level under each reading ability indicator, may include:

[0076] Using pre-configured weights corresponding to each reading ability indicator, the scores of each reading ability indicator are linearly weighted, and the result is used as the total reading ability level of the test subject.

[0077] Further refer to Table 1 below, which illustrates, under the condition that the specified assessment materials include passage reading and rapid number naming, the following types of eye movement features are extracted from the time dimension: several types of eye movement features contained in the time feature set 1 corresponding to reading comprehension, several types of eye movement features contained in the time feature set 2 corresponding to reading attention, and several types of eye movement features contained in the time feature set 3 corresponding to reading memory.

[0078]

[0079]

[0080] Table 1 above only illustrates some types of eye movement features in the time dimension and does not exhaust all types of eye movement features under each reading ability assessment indicator.

[0081] Further refer to Table 2 below, which illustrates, under the condition that the specified assessment materials include passage reading and rapid number naming, the following types of eye movement features are extracted from the spatial dimension: several types of eye movement features contained in spatial feature set 1 corresponding to reading comprehension, several types of eye movement features contained in spatial feature set 2 corresponding to reading attention, and several types of eye movement features contained in spatial feature set 3 corresponding to reading memory.

[0082]

[0083]

[0084] Table 2 above only illustrates some types of eye movement features in the spatial dimension and does not exhaust all types of eye movement features under each reading ability assessment indicator.

[0085] In some embodiments of this application, the impact of the completeness of the assessment process on the overall reading ability level of the test takers is further considered, and the overall reading ability level of the test takers is adjusted accordingly.

[0086] Specifically, the completeness of the test can be measured by the number of regions of interest and other regions browsed by the test taker during the process of reading the specified assessment materials, as well as the time spent. This embodiment provides an optional implementation method as follows:

[0087] First, we calculate the ratio of regions of interest containing fixation points to the total number of regions of interest in each region of interest on the specified assessment material, and the total time spent on fixation points within all regions of interest.

[0088] Specifically, when reading the designated assessment material, the test taker's gaze will pass through regions of interest (ROIs) or other regions. If the number of times the test taker's gaze falls on ROIs during the entire reading process is too small, and the total time spent reading ROIs is too low, it indicates that the test's completeness is insufficient, and the test taker's overall reading ability level calculated above can be penalized. Therefore, in this embodiment, the ratio of the number of ROIs containing gaze points to the total number of ROIs, and the total time spent on all gaze points within ROIs, can be calculated.

[0089] In this embodiment, any one or a combination of the above two indicators, namely the proportion of the region of interest and the total time spent on fixation within the region of interest, can be used as an indicator to measure the completeness of the implementation. Based on this, and according to the relationship between the indicators and the preset indicator thresholds, it is determined whether to penalize the test subject's overall reading ability level.

[0090] For example, if the ratio of the number of regions of interest containing fixations to the total number of regions of interest is lower than a set ratio threshold (e.g., 0.55 or other values), and / or the total time spent on fixations within all regions of interest is lower than a set duration threshold (e.g., 10s or other values), it indicates that the implementation is not complete enough and the test subject's overall reading ability level needs to be penalized. For example, for the test subject's overall reading ability level in the form of a numerical score, it can be multiplied by a preset penalty weight (e.g., 0.3 or other values), and the result can be used as the test subject's penalized overall reading ability level.

[0091] Of course, if the ratio of the number of regions of interest containing fixation points to the total number of regions of interest is not less than the set ratio threshold, and / or the total time spent on fixation points within all regions of interest is not less than the set duration threshold, then the implementation integrity requirement is met, and there is no need to penalize the overall reading ability level. In other words, the overall reading ability level of the test subject obtained above is taken as the final result.

[0092] In some embodiments of this application, the process of performing fixation point segmentation based on the original eye movement data to obtain a set of fixation points in step S110 of the foregoing embodiments is further described.

[0093] Specifically, the process can be handled as follows:

[0094] S1. Preprocess the raw eye movement data to obtain preprocessed eye movement data.

[0095] The preprocessing of raw eye-tracking data can include the following preprocessing operations:

[0096] S11, Abnormal data cleaning.

[0097] Considering that test subjects exhibit normal saccade behavior during the test, a reasonable maximum interval for saccade behavior can be set in this case, such as 75ms or other optional values.

[0098] Based on this, if the duration of the missing data segment in the collected raw eye movement data exceeds the above-mentioned maximum interval time, it is considered not to be a reasonable eye saccade behavior of the test subject. It may be that the eyes are off the screen, resulting in the inability to collect data for the corresponding time period. Therefore, the missing data segment can be left unprocessed.

[0099] If the duration of the missing data segment in the collected raw eye movement data does not exceed the maximum interval time mentioned above, the missing data segment can be filled in by linear interpolation based on the raw eye movement data before and after the missing data segment.

[0100] S12, Left and right eye data selection.

[0101] Specifically, the eye-tracking data cleaned of the abnormal data includes trajectory data for each of the left and right eyes. Therefore, the subject's gaze point on the screen can be determined based on the trajectory data of the left and right eyes.

[0102] 1) When there is a missing data point for the left eye but a valid data point for the right eye, the right eye data can be selected as the eye movement data for that time point.

[0103] 2) When there is a missing right eye data but a valid left eye data point, the left eye data can be selected as the eye movement data for that point.

[0104] 3) For time points where both left and right eye data are valid, the average data of the left and right eyes can be selected as the eye movement data for that time point.

[0105] 4) For time points where data for both eyes is missing, the eye movement data at that point can be considered invalid and the value should be set to empty.

[0106] S13, Median noise reduction filter.

[0107] To smooth the eye-tracking data, this embodiment can perform median noise reduction filtering. Specifically, n sampling points can be selected as a time window, and median filtering can be performed on each eye-tracking data point in the eye-tracking data sequentially:

[0108] 1) For the x-axis data of eye tracking data points, starting from the (n+1) / 2th sampled eye tracking data point, its value can be replaced with the median of the x-axis data of the (n-1) / 2th eye tracking data points before and after it.

[0109] Taking n as an example, starting from the 4th sampled eye movement data point, its value is replaced with the median of the x-axis data of the first 3 and last 3 eye movement data points.

[0110] 2) Similarly, for the y-axis data of eye-tracking data points, starting from the (n+1) / 2th sampled eye-tracking data point, its value can be replaced with the median of the y-axis data of the (n-1) / 2th eye-tracking data points before and after it.

[0111] S2. For each eye movement data point in the preprocessed eye movement data, calculate the average angular velocity between the preceding and following eye movement data points, and use it as the instantaneous angular velocity of the current eye movement data point.

[0112] Specifically, in this embodiment, fixation points and saccade points are divided according to the magnitude of the instantaneous angular velocity of the eye movement data points. Therefore, it is necessary to determine the corresponding instantaneous angular velocity for each eye movement data point. This instantaneous angular velocity can represent the instantaneous angular velocity of the subject's eye as it moves past the current eye movement data point.

[0113] In this embodiment, to reduce errors, the average angular velocity between the first and last eye movement data points is used as the instantaneous angular velocity of the middle eye movement data point. Taking the second eye movement data point as an example, the average angular velocity between the first and third eye movement data points can be calculated as the instantaneous angular velocity of the second eye movement data point.

[0114] S3. Eye movement data points whose instantaneous angular velocity exceeds the set angular velocity threshold are classified as saccade points, and eye movement data points whose instantaneous angular velocity does not exceed the set angular velocity threshold are classified as fixation points, thus obtaining a set of fixation points.

[0115] Specifically, in the previous step, the instantaneous angular velocity of each eye movement data point in the eye movement data was calculated. In this step, eye movement data points whose instantaneous angular velocity exceeds the set angular velocity threshold are classified as saccade points, and the rest are classified as fixation points, thus obtaining the set of fixation points.

[0116] The angular velocity threshold can be set to 30 rad / s or other optional values.

[0117] Alternatively, the saccade points and fixation points divided in step S3 above can be further filtered.

[0118] For example, after dividing saccades and fixation points according to instantaneous angular velocity in step S3, saccade segments are formed by consecutive saccades, and fixation segments are formed by consecutive fixation points. Thus, eye movement data is divided into several saccade segments and fixation segments.

[0119] For each eye jump segment:

[0120] The angle and time interval between the last fixation point in a fixation segment preceding the saccade and the first fixation point in a fixation segment following the saccade can be calculated. If the angle is lower than a set angle threshold and the time interval is less than a first time threshold, it indicates that the saccade is too short and is a misclassification. Therefore, the saccade is merged with the preceding and following fixation segments, that is, all points in the saccade are changed to fixation points.

[0121] The aforementioned angle threshold can be set to 0.5 degrees or other values, and the first time threshold can be set to 75ms or other values.

[0122] For each fixation segment:

[0123] The time interval between the first and last fixation points within the fixation segment can be calculated. If the time interval is less than the second time threshold, it indicates that the fixation segment is too short and is a misclassification. Therefore, each fixation point in the fixation segment can be reclassified as a saccade point.

[0124] The second time threshold can be 60ms or other values.

[0125] After processing the saccade and fixation segments as described above, the final set of fixation points is formed from the resulting fixation points, which makes the division of fixation points more accurate.

[0126] Reference Figure 2a and Figure 2b In the example, the gray dots represent fixation points, and the lines connecting the gray dots indicate the order of different fixation points. Of course, the area of ​​the gray dots in the figure may vary. One gray dot can represent a fixation segment composed of several consecutive fixation points. The center of the gray dot is the average position of the fixation points in the corresponding fixation segment. The size of the gray dot represents the total time spent by all fixation points in the corresponding fixation segment; the longer the time, the larger the area of ​​the gray dot.

[0127] For step S2 above, for each eye movement data point in the preprocessed eye movement data, the process of calculating the average angular velocity between one eye movement data point before and after the previous eye movement data point as the instantaneous angular velocity of the current eye movement data point, the embodiments of this application provide several different calculation methods.

[0128] For ease of description, we define the current eye movement data point as p, the previous eye movement data point as s, and the next eye movement data point as e. When collecting eye movement data s, the subject's eyes are located at position o1. When collecting eye movement data p, the subject's eyes are located at position o2. When collecting eye movement data e, the subject's eyes are located at position o3. The center point of the screen is c.

[0129] The first type

[0130] Combination Figure 4 As shown:

[0131] For the current eye movement data point p, when calculating the average angular velocity L” between one eye movement data point before and after:

[0132] Based on eye movement data point s and the position o1 of the subject's eyes when collecting eye movement data s, the first vector pointing from point o1 to point s can be calculated.

[0133] Similarly, based on eye movement data point e and the subject's eye position o3 when collecting eye movement data e, a second vector pointing from point o3 to point e can be calculated.

[0134] If we ignore the changes in the subject's eye position when collecting different eye movement data points, that is, if we assume that the subject's eye position remains unchanged when collecting eye movement data points s, p, and e, then points o1, o2, and o3 coincide. Therefore, we can calculate the angle θ between the two vectors based on the first and second vectors mentioned above.

[0135] Furthermore, the target time interval t between eye-tracking data point s and eye-tracking data point e is calculated.

[0136] Calculate the average angular velocity L” = θ” / t

[0137] Of course, the above method calculates the average angular velocity L between one eye movement data point before and after a given point, ignoring changes in the subject's eye position when collecting different eye movement data points. However, in actual testing, the subject's eye position may undergo subtle changes, such as... Figure 4 As shown, points o1, o2, and o3 are not exactly the same. In this case, the angle between the first vector and the second vector is actually the angle at the intersection point o” of the extension of the line segment from point s to point o1 and the extension of the line segment from point e to point o3, and is not an accurate angle value.

[0138] To further improve the accuracy of the calculated average angular velocity L”, this application provides another calculation method, as follows:

[0139] The second type

[0140] Combination Figure 4 As shown:

[0141] S21. Calculate the target time interval t between the previous eye movement data point s and the next eye movement data point e.

[0142] S22. Calculate the distances sc and ec from the previous eye-tracking data point s and the next eye-tracking data point e to the center point c of the screen, respectively, and the distance se between the previous eye-tracking data point s and the next eye-tracking data point e.

[0143] S23. Obtain the vertical distance d from the subject's eye to the screen when collecting the current eye movement data point p.

[0144] S24. Based on the geometric relationship between the previous eye movement data point s, the next eye movement data point e, the center point c of the screen, and the target position point o2 where the subject's eyes are located when the current eye movement data point is collected, calculate the target angle value θ of the angle formed by the triangle formed by the previous eye movement data point s, the next eye movement data point e, and the target position point o2.

[0145] Specifically, based on geometric relationships, the triangle formed by points s, c, and o2 is a right triangle, and the angle at point c is a right angle. Therefore, given the lengths of the two legs sc and d, the length of the hypotenuse can be calculated, which gives the distance so2 between points s and o2.

[0146] so2 = math.sqrt(sc^2 + d^2)

[0147] Similarly, the triangle formed by points e, c, and o2 is a right triangle, and the angle at point c is a right angle. Therefore, given the lengths of the two legs, ec and d, we can calculate the length of the hypotenuse, which is the distance eo2 between points e and o2.

[0148] eo2 = math.sqrt(ec^2 + d^2)

[0149] In the triangle formed by points s, e, and o2, given the lengths of the three sides so2, eo2, and se, the target angle θ of the included angle containing point o2 can be calculated.

[0150] S25. Divide the target angle value θ by the target time interval t, and use the result as the average angular velocity L between one eye movement data point before and after the current eye movement data point.

[0151] L=θ / t

[0152] It should be noted that the calculation method provided in this embodiment requires calculating the instantaneous angular velocity of the intermediate eye movement data point based on the data of the two preceding and following eye movement data points. For the first and last eye movement data points in the preprocessed eye movement data, since the preceding or following eye movement data point is missing, it is impossible to obtain the corresponding instantaneous angular velocity according to the calculation method of this embodiment. Therefore, in this embodiment, the instantaneous angular velocities of the first and last eye movement data points can be set to a default value, such as a null value or a default non-null value, or they can be set to the instantaneous angular velocity of the nearest adjacent eye movement data point.

[0153] It should be further noted that, since there may be missing data segments in the preprocessed eye-tracking data, some eye-tracking data points may also be missing the preceding or following eye-tracking data point. For such eye-tracking data points, the corresponding instantaneous angular velocity value can also be set in the manner mentioned in the previous paragraph.

[0154] In some embodiments of this application, a practical application scenario for a reading ability assessment method is further described.

[0155] In the application scenario described in this embodiment, by adopting the reading ability assessment method described in the foregoing embodiment, the overall reading ability level of the test subject can be obtained. Based on this, the overall reading ability level of the test subject can be output and displayed, and the level of the test subject under each reading ability indicator can be displayed in the form of statistical charts.

[0156] like Figure 5 As shown:

[0157] The reading ability assessment interface can display the test taker's personal information, such as name, gender, grade, date of birth, and vision information. In addition, it can also display the assessment time.

[0158] The interface can also display the test taker's overall reading ability level, specifically in the form of a score (such as displaying the score value under a 100-point system) or a rating system (such as excellent, good, average, need improvement).

[0159] In addition, to more intuitively show test takers their levels across different reading ability indicators, statistical charts can be used to display their levels for each indicator. Figure 5 Specifically, taking reading comprehension, reading attention, and reading memory as examples, we will display them in the form of a radar chart. In addition, they can also be displayed in the form of bar charts, pie charts, tables, etc.

[0160] Of course, to help test subjects understand their own level in each indicator, reference values ​​for each indicator can also be displayed simultaneously. These reference values ​​can be the average or minimum level of the normal user group for each indicator.

[0161] Furthermore, the gaze trajectory diagram formed by the test subject reading the specified assessment material can be displayed simultaneously, with the specified assessment material superimposed on the gaze trajectory diagram, such as... Figure 5 As shown on the right. Figure 5 The example includes two assessment materials: passage reading and quick naming. By switching between the two assessment materials, the gaze trajectory diagram of the corresponding assessment material can be displayed in the area below. The gaze trajectory diagram can be played in the form of a video animation, and displayed in the order of each gaze point.

[0162] Furthermore, based on the foregoing, each reading ability indicator contains several types of eye movement features. The method provided in this embodiment can also support the function of displaying each type of eye movement feature under each reading ability indicator. That is, this application can respond to the user's instruction to request eye movement features under the target reading ability indicator and display several types of eye movement features under the target reading ability indicator in the form of statistical charts.

[0163] The displayed eye movement features can be any type of eye movement feature under the target reading ability indicator, or it can be any of the default types of eye movement features.

[0164] Further reference Figure 6 As shown in this embodiment, the top N types of eye movement features with the highest discriminative power for normal reading ability samples and high-risk reading disorder samples can also be statistically analyzed across all reading ability indicators. These top N types of eye movement features are then displayed in the form of statistical charts.

[0165] The reading ability assessment device provided in the embodiments of this application is described below. The reading ability assessment device described below can be referred to in correspondence with the reading ability assessment method described above.

[0166] See Figure 7 , Figure 7 This is a schematic diagram of the structure of a reading ability assessment device disclosed in an embodiment of this application.

[0167] like Figure 7 As shown, the device may include:

[0168] The raw eye movement data acquisition unit 11 is used to acquire the raw eye movement data of the test subject during the process of reading the specified assessment material;

[0169] The fixation point segmentation unit 12 is used to segment fixation points based on the original eye movement data to obtain a set of fixation points.

[0170] The eye movement feature acquisition unit 13 is used to acquire eye movement features corresponding to each pre-configured reading ability indicator based on the region of interest divided on the specified assessment material and the set of fixation points. The reading ability indicators include at least one of reading comprehension, reading attention, and reading memory.

[0171] The indicator level determination unit 14 is used to determine the level of the test subject under the corresponding reading ability indicator based on the eye movement characteristics corresponding to each reading ability indicator;

[0172] The comprehensive level determination unit 15 is used to determine the comprehensive reading ability level of the test subject based on the test subject's level under each reading ability indicator.

[0173] Optionally, if the eye movement features corresponding to each reading ability indicator include at least one type of eye movement feature, then the process by which the indicator level determination unit determines the test subject's level under the corresponding reading ability indicator based on the eye movement features corresponding to each reading ability indicator includes:

[0174] Using pre-configured weights corresponding to each type of eye movement feature, the eye movement features corresponding to each reading ability indicator are linearly weighted to obtain the score of each reading ability indicator, which is used as the test subject's level under the corresponding reading ability indicator.

[0175] The process by which the aforementioned comprehensive level determination unit determines the overall reading ability level of the test subject based on the test subject's level under each reading ability indicator includes:

[0176] Using pre-configured weights corresponding to each reading ability indicator, the scores of each reading ability indicator are linearly weighted, and the result is used as the total reading ability level of the test subject.

[0177] Optionally, the process by which the eye-tracking feature acquisition unit acquires eye-tracking features corresponding to each pre-configured reading ability indicator based on the region of interest divided on the specified assessment material and the set of fixations includes:

[0178] The content area of ​​the specified evaluation material is divided into several regions of interest according to the set division rules;

[0179] Based on the position of each gaze point in the set of gaze points, determine the region of interest to which each gaze point belongs. For gaze points outside the region of interest, mark them as belonging to other regions.

[0180] Based on the region of interest and the position and time attributes of each fixation point in the set of fixations, eye movement features corresponding to each pre-configured reading ability index are obtained.

[0181] Optionally, the apparatus of this application may further include:

[0182] The level penalty unit is used to calculate the ratio of the number of regions of interest containing fixation points to the total number of regions of interest in each region of interest on the specified assessment material, and the total time spent on fixation points within all regions of interest; if the ratio is lower than a set ratio threshold, and / or the total time spent is lower than a set duration threshold, the overall reading ability level of the test subject is penalized to obtain the overall reading ability level of the test subject after penalty.

[0183] Optionally, the process by which the above-mentioned gaze point segmentation unit performs gaze point segmentation based on the original eye movement data to obtain a gaze point set includes:

[0184] The raw eye movement data is preprocessed to obtain preprocessed eye movement data;

[0185] For each eye movement data point in the preprocessed eye movement data, the average angular velocity between the preceding and following eye movement data points is calculated and used as the instantaneous angular velocity of the current eye movement data point;

[0186] Eye movement data points whose instantaneous angular velocity exceeds a set angular velocity threshold are classified as saccade points, and eye movement data points whose instantaneous angular velocity does not exceed the set angular velocity threshold are classified as fixation points, thus obtaining a set of fixation points.

[0187] Optionally, the process by which the above-mentioned gaze point segmentation unit calculates the average angular velocity between one eye movement data point before and after the current eye movement data point includes:

[0188] Calculate the target time interval between the previous eye movement data point and the next eye movement data point;

[0189] Calculate the distance from the previous eye movement data point and the next eye movement data point to the center of the screen, as well as the distance between the previous eye movement data point and the next eye movement data point;

[0190] The vertical distance from the subject's eye to the screen is obtained when the current eye movement data point is collected;

[0191] Based on the geometric relationship between the previous eye movement data point, the next eye movement data point, the center point of the screen, and the target position point where the subject's eyes are located when the current eye movement data point is collected, the target angle value of the angle formed by the previous eye movement data point, the next eye movement data point, and the target position point is calculated.

[0192] Divide the target angle value by the target time interval, and the result is used as the average angular velocity between the two eye movement data points before and after the current eye movement data point.

[0193] Optionally, the process by which the above-mentioned gaze point segmentation unit calculates the target angle value based on the geometric relationship between the previous eye movement data point, the next eye movement data point, the screen center point, and the target position point where the subject's eyes are located when the current eye movement data point is collected, includes:

[0194] In the right triangle formed by the previous eye movement data point, the center point of the screen, and the target position point where the subject's eyes were located when the current eye movement data point was acquired, calculate the distance from the previous eye movement data point to the target position point, and...

[0195] In the right triangle formed by the next eye movement data point, the center point of the screen, and the target position point where the subject's eyes were located when the current eye movement data point was collected, calculate the distance from the next eye movement data point to the target position point.

[0196] Calculate the target angle value of the included angle of the target location point within the triangle formed by the previous eye movement data point, the next eye movement data point, and the target location point.

[0197] Optionally, the process by which the above-mentioned gaze point segmentation unit performs gaze point segmentation based on the original eye movement data to obtain a gaze point set further includes:

[0198] After dividing saccade points and fixation points according to instantaneous angular velocity, saccade segments are composed of consecutive saccade points, and fixation segments are composed of consecutive fixation points.

[0199] For each eye jump segment:

[0200] Calculate the angle and time interval between the last fixation point in the fixation segment preceding the saccade and the first fixation point in the fixation segment following the saccade. If the angle is lower than a set angle threshold and the time interval is less than a first time threshold, then merge the saccade with the preceding and following fixation segments.

[0201] For each fixation segment:

[0202] Calculate the time interval between the first fixation point and the last fixation point within the fixation segment. If the time interval is less than the second time threshold, then reclassify each fixation point in the fixation segment as a saccade point.

[0203] The final set of fixations is composed of all the fixations obtained.

[0204] Optionally, the apparatus of this application may further include:

[0205] The first display unit is used to display the test subject's overall reading ability level, and to display the test subject's level under each reading ability indicator in the form of statistical charts.

[0206] Optionally, the apparatus of this application may further include:

[0207] The second display unit is used to display the gaze trajectory diagram of the test subject, on which the specified assessment material is superimposed.

[0208] Optionally, the apparatus of this application may further include:

[0209] The third display unit is used to respond to the user's instruction to request eye movement features under the target reading ability index, and to display several types of eye movement features under the target reading ability index in the form of statistical charts.

[0210] The reading ability assessment device provided in this application embodiment can be applied to reading ability assessment devices, such as terminals: mobile phones, computers, etc. Optionally, Figure 8 The hardware structure block diagram of the reading ability assessment device is shown below. Figure 8 The hardware structure of a reading ability assessment device may include: at least one processor 1, at least one communication interface 2, at least one memory 3, and at least one communication bus 4;

[0211] In this embodiment of the application, the number of processor 1, communication interface 2, memory 3, and communication bus 4 is at least one, and processor 1, communication interface 2, and memory 3 communicate with each other through communication bus 4;

[0212] Processor 1 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention.

[0213] Memory 3 may include high-speed RAM, and may also include non-volatile memory, such as at least one disk storage device;

[0214] The memory stores a program, which the processor can call to execute the various steps of the reading ability assessment method described above.

[0215] This application embodiment also provides a storage medium that can store a program suitable for execution by a processor, the program being used to execute each step of the above-described reading ability assessment method.

[0216] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0217] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0218] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A reading ability assessment method characterized by, include: Obtain raw eye movement data of the test subject during the process of reading the specified assessment materials; Based on the original eye movement data, fixation points are segmented to obtain a set of fixation points; Based on the regions of interest and the set of fixations defined on the specified assessment material, eye movement features corresponding to each pre-configured reading ability indicator are obtained. The reading ability indicators include at least one of reading comprehension, reading attention, and reading memory. Based on the eye movement characteristics corresponding to each reading ability indicator, the test subject's level under the corresponding reading ability indicator is determined, and based on the test subject's level under each reading ability indicator, the test subject's comprehensive reading ability level is determined. The process of segmenting fixation points based on the original eye-tracking data to obtain a set of fixation points includes: Based on each eye movement data point in the original eye movement data, the average angular velocity between the preceding and following eye movement data points is calculated as the instantaneous angular velocity of the current eye movement data point. The process of calculating the average angular velocity between the preceding and following eye movement data points for the current eye movement data point includes: calculating the target time interval between the preceding and following eye movement data points; calculating the distance from each of the preceding and following eye movement data points to the center point of the screen, and the distance between the preceding and following eye movement data points; and acquiring the collected data. When collecting eye-tracking data points, the vertical distance from the subject's eyes to the screen is measured. Based on the geometric relationship between the previous eye-tracking data point, the next eye-tracking data point, the center point of the screen, and the target position point where the subject's eyes are located when the current eye-tracking data point is collected, the target angle value of the angle formed by the triangle formed by the previous eye-tracking data point, the next eye-tracking data point, and the target position point is calculated. The target angle value is divided by the target time interval, and the result is used as the average angular velocity between the two eye-tracking data points before and after the current eye-tracking data point. Eye movement data points whose instantaneous angular velocity exceeds a set angular velocity threshold are classified as saccade points, and eye movement data points whose instantaneous angular velocity does not exceed the set angular velocity threshold are classified as fixation points, thus obtaining a set of fixation points.

2. The method of claim 1, wherein, The reading ability indicators include: reading comprehension, reading attention, and reading memory; The eye movement features corresponding to each reading ability indicator include: the first eye movement feature corresponding to the reading comprehension indicator, the second eye movement feature corresponding to the reading attention indicator, and the third eye movement feature corresponding to the reading memory indicator. Based on the eye movement characteristics corresponding to each reading ability indicator, the test subject's level under the corresponding reading ability indicator is determined, and based on the test subject's level under each reading ability indicator, the test subject's overall reading ability level is determined, including: Based on the first eye movement feature, the test subject's reading comprehension level is determined; based on the second eye movement feature, the test subject's reading attention level is determined; and based on the third eye movement feature, the test subject's reading memory level is determined. Based on the test subject's reading comprehension level, reading attention level, and reading memory level, the test subject's overall reading ability level is determined.

3. The method of claim 1, wherein, Each reading ability indicator corresponds to at least one type of eye movement feature; The process of determining the test subject's level under each reading ability indicator based on the eye movement characteristics corresponding to that indicator includes: Using pre-configured weights corresponding to each type of eye movement feature, the eye movement features corresponding to each reading ability indicator are linearly weighted to obtain the score of each reading ability indicator, which is used as the test subject's level under the corresponding reading ability indicator. The process of determining the overall reading ability level of the test subject based on their level under each reading ability indicator includes: Using pre-configured weights corresponding to each reading ability indicator, the scores of each reading ability indicator are linearly weighted, and the result is used as the overall reading ability level of the test subject.

4. The method according to claim 3, characterized in that, The pre-configured weights corresponding to each type of eye movement feature and each reading ability indicator were obtained by analyzing training sample data labeled with comprehensive reading ability levels using exploratory factor analysis.

5. The method according to claim 3, characterized in that, The eye movement characteristics corresponding to each reading ability indicator include: Eye movement features in the time dimension, and eye movement features in the spatial dimension; Furthermore, each dimension of eye movement features includes at least one type of eye movement feature.

6. The method according to claim 1, characterized in that, Based on the regions of interest defined on the specified assessment material and the set of fixations, eye movement features corresponding to each pre-configured reading ability indicator are obtained, including: The content area of ​​the specified evaluation material is divided into several regions of interest according to the set division rules; Based on the position of each gaze point in the set of gaze points, determine the region of interest to which each gaze point belongs. For gaze points outside the region of interest, mark them as belonging to other regions. Based on the region of interest and the position and time attributes of each fixation point in the set of fixations, eye movement features corresponding to each pre-configured reading ability index are obtained.

7. The method according to claim 6, characterized in that, Also includes: The ratio of the number of regions of interest containing fixation points to the total number of regions of interest in each region of interest on the specified evaluation material is calculated, as well as the total time spent on fixation points within all regions of interest. If the ratio is lower than a set ratio threshold, and / or the total time consumed is lower than a set duration threshold, the test subject's overall reading ability level is penalized to obtain the test subject's penalized overall reading ability level.

8. The method according to claim 1, characterized in that, The process of calculating the average angular velocity between one eye movement data point and one eye movement data point before and after each eye movement data point in the original eye movement data includes: The raw eye movement data is preprocessed to obtain preprocessed eye movement data; For each eye movement data point in the preprocessed eye movement data, calculate the average angular velocity between one eye movement data point before and after it.

9. The method according to claim 1, characterized in that, The process of calculating the target angle value based on the geometric relationship between the previous eye movement data point, the next eye movement data point, the center point of the screen, and the target position point where the subject's eyes were located when the current eye movement data point was collected includes: In the right triangle formed by the previous eye movement data point, the center point of the screen, and the target position point where the subject's eyes were located when the current eye movement data point was acquired, calculate the distance from the previous eye movement data point to the target position point, and... In the right triangle formed by the next eye movement data point, the center point of the screen, and the target position point where the subject's eyes were located when the current eye movement data point was collected, calculate the distance from the next eye movement data point to the target position point. Calculate the target angle value of the included angle of the target location point within the triangle formed by the previous eye movement data point, the next eye movement data point, and the target location point.

10. The method according to claim 8, characterized in that, The process of segmenting fixation points based on the original eye-tracking data to obtain a set of fixation points also includes: After dividing saccade points and fixation points according to instantaneous angular velocity, saccade segments are composed of consecutive saccade points, and fixation segments are composed of consecutive fixation points. For each eye jump segment: Calculate the angle and time interval between the last fixation point in the fixation segment preceding the saccade and the first fixation point in the fixation segment following the saccade. If the angle is lower than a set angle threshold and the time interval is less than a first time threshold, then merge the saccade with the preceding and following fixation segments. For each fixation segment: Calculate the time interval between the first fixation point and the last fixation point within the fixation segment. If the time interval is less than the second time threshold, then reclassify each fixation point in the fixation segment as a saccade point. The final set of fixations is composed of all the fixations obtained.

11. The method according to any one of claims 1-10, characterized in that, Also includes: The test subject's overall reading ability level is displayed, and the test subject's level under each reading ability indicator is displayed in the form of statistical charts.

12. The method according to claim 11, characterized in that, Also includes: The gaze trajectory diagram of the test subject is displayed, and the specified assessment material is superimposed on the gaze trajectory diagram.

13. The method according to claim 11, characterized in that, Also includes: In response to a user's request for eye movement features under a target reading ability indicator, several types of eye movement features under the target reading ability indicator are displayed in the form of statistical charts.

14. A reading ability assessment device, characterized in that, include: The raw eye movement data acquisition unit is used to acquire raw eye movement data of the test subject during the process of reading the specified assessment materials; A gaze point segmentation unit is used to segment gaze points based on the original eye movement data to obtain a set of gaze points; An eye movement feature acquisition unit is used to acquire eye movement features corresponding to each pre-configured reading ability indicator based on the region of interest divided on the specified assessment material and the set of fixation points. The reading ability indicators include at least one of reading comprehension, reading attention, and reading memory. The indicator level determination unit is used to determine the level of the test subject under the corresponding reading ability indicator based on the eye movement characteristics corresponding to each reading ability indicator; The comprehensive level determination unit is used to determine the comprehensive reading ability level of the test subject based on the test subject's level under each reading ability indicator; Specifically, the gaze point segmentation unit is used for: Based on each eye movement data point in the original eye movement data, the average angular velocity between the preceding and following eye movement data points is calculated as the instantaneous angular velocity of the current eye movement data point. The process of calculating the average angular velocity between the preceding and following eye movement data points for the current eye movement data point includes: calculating the target time interval between the preceding and following eye movement data points; calculating the distance from each of the preceding and following eye movement data points to the center point of the screen, and the distance between the preceding and following eye movement data points; and acquiring the collected data. When collecting eye-tracking data points, the vertical distance from the subject's eyes to the screen is measured. Based on the geometric relationship between the previous eye-tracking data point, the next eye-tracking data point, the center point of the screen, and the target position point where the subject's eyes are located when the current eye-tracking data point is collected, the target angle value of the angle formed by the triangle formed by the previous eye-tracking data point, the next eye-tracking data point, and the target position point is calculated. The target angle value is divided by the target time interval, and the result is used as the average angular velocity between the two eye-tracking data points before and after the current eye-tracking data point. Eye movement data points whose instantaneous angular velocity exceeds a set angular velocity threshold are classified as saccade points, and eye movement data points whose instantaneous angular velocity does not exceed the set angular velocity threshold are classified as fixation points, thus obtaining a set of fixation points.

15. A reading ability assessment device, characterized in that, include: Memory and processor; The memory is used to store programs; The processor is configured to execute the program to implement each step of the reading ability assessment method as described in any one of claims 1 to 13.

16. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements each step of the reading ability assessment method as described in any one of claims 1 to 13.

Images