Sight tracking-based ideological and political large-scale course teaching attention assessment data acquisition method

Abstract

The invention discloses an ideological and political large-scale course teaching attention assessment data acquisition method based on sight tracking. The method comprises the following steps: S1, high-definition large-scene eye movement tracking photographing equipment captures eye images of students in a course teaching process in real time through an infrared camera; s2, performing pupil area detection and pupil center fitting on the eye image to obtain sight tracking data; s3, performing gaze hotspot analysis and gaze duration statistics according to the sight tracking data to quantify attention distribution and change trend of the students and visually display the attention distribution and change trend; and S4, generating a teaching quality evaluation report and teaching optimization suggestions according to the gazing hotspot analysis and gazing duration statistical results. The classroom teaching effect can be objectively evaluated in real time, subjective evaluation errors are reduced, the accuracy and efficiency of teaching feedback are improved, the method is suitable for various scenes such as traditional classrooms, online education and experiment teaching, and the method has the industrial application prospect of education digital transformation.

Description

Technical Field

[0001] The present invention belongs to the field of gaze tracking technology, and in particular relates to a method for collecting data on attention assessment in large-scale ideological and political courses based on gaze tracking. Background Art

[0002] Traditional methods for assessing teaching attention rely primarily on methods such as teacher classroom observation, indirect student questionnaires, or indirect analysis of student performance. However, manual observation requires evaluators to record classroom activities over a long period of time, which is not only labor-intensive but also difficult to achieve comprehensive attention for all students. Questionnaires are subject to significant subjective factors, and student feedback may not be accurate or comprehensive. Furthermore, performance analysis and other methods are indirect and lack accurate analysis. Feedback from these methods is often delayed, making it difficult to provide timely and effective support for teaching improvements.

[0003] Existing techniques use classroom video analysis to assess teaching quality, using video playback to analyze teachers' teaching methods and students' attention and performance. However, this method primarily focuses on teacher behavior and lacks quantitative assessment of students' cognitive states, making it difficult to accurately determine students' attention distribution and learning engagement.

[0004] Gaze tracking technology, which uses eye movements to infer an individual's focus of attention, has been widely used in fields such as psychology, market research, and human-computer interaction. This technology can record an individual's gaze point, gaze trajectory, and gaze duration in real time, thereby inferring their interests and information processing patterns.

[0005] However, gaze tracking technology has not yet been systematically applied to the dynamic evaluation of teaching scenarios. Especially in large-scale ideological and political classroom environments where the teaching content is relatively abstract, there are still many challenges in accurately and stably tracking students' gazes and converting them into quantifiable teaching quality evaluation indicators. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to address the deficiencies of the above-mentioned existing technologies and provide a data collection method for attention assessment in large-scale ideological and political courses based on eye tracking. By using high-precision eye tracking equipment to collect students' eye gaze data in real time, the students' attention level can be quantified, and an objective analysis of students' attention in class can be achieved. This method is helpful for analyzing and evaluating the teaching quality of large-scale ideological and political courses and providing a scientific basis for teaching optimization.

[0007] In order to achieve the above technical objectives, the technical solution adopted by the present invention is:

[0008] The method for collecting data on attention assessment in large-scale ideological and political courses based on eye tracking includes the following steps:

[0009] S1, high-definition large-scene eye tracking photography equipment uses infrared cameras to capture students' eye images in real time during the course teaching process;

[0010] S2. Perform pupil area detection and pupil center fitting on the eye image to obtain gaze tracking data;

[0011] S3. Perform gaze hotspot analysis and gaze duration statistics based on gaze tracking data to quantify students’ attention distribution and changing trends and visualize them.

[0012] S4. Generate a teaching quality evaluation report and teaching optimization suggestions based on the gaze hotspot analysis and gaze duration statistics.

[0013] To optimize the above technical solutions, specific measures taken also include:

[0014] The above-mentioned S2 performs pupil area detection and pupil center fitting on the eye image to obtain eye movement data, including:

[0015] S21, converting the eye image into a grayscale image, and removing noise using a Gaussian filter kernel function to obtain a denoised grayscale image G(x,y);

[0016] S22. Calculate the horizontal and vertical gradient values ​​of the denoised grayscale image G(x,y) to obtain a gradient amplitude G, and automatically determine an optimal binarization threshold using the Otsu threshold segmentation method. Identify areas with a gradient amplitude greater than the threshold as pupil areas.

[0017] S23. Use an ellipse fitting algorithm to perform edge detection on the pupil area, and obtain the geometric parameters of the pupil based on the ellipse model fitting, so as to determine the center position and shape of the pupil, and finally obtain the gaze tracking data.

[0018] The Gaussian filter kernel function described in S21 above is as follows:

[0019]

[0020] Where G(x,y) is the denoised grayscale image, x and y are the horizontal and vertical coordinates of the grayscale image pixels, and σ is the smoothness control parameter.

[0021] The gradient calculation formula in the above S22 is as follows:

[0022]

[0023] Among them, G x and G y are the gradient values ​​of G(x,y) in the horizontal and vertical directions respectively.

[0024] The ellipse fitting algorithm described in S23 above specifically obtains the parameters of the ellipse by solving the following objective function, and the corresponding ellipse center is the pupil center:

[0025]

[0026] Where f(A,B,C,D,E,F) is the ellipse parameter set, A,B,C,D,E,F are the ellipse parameters, (x,y) are the edge point coordinates, and (x0,y0) are the pupil center coordinates.

[0027] The aforementioned S3 performs gaze hotspot analysis and gaze duration statistics based on gaze tracking data to quantify students' attention distribution and changing trends and visualize them, including:

[0028] S31. Determine the gaze area based on the gaze tracking data, and use Gaussian kernel density estimation to calculate the gaze density f(X,Y) of different areas. Cluster the gaze densities of different gaze areas to generate a classroom attention heat map to intuitively present the distribution of students' attention.

[0029] S32. Determine the student's gaze area based on the gaze tracking data and identify effective gazes based on the gaze duration. Calculate the attention stability index based on the number of effective gazes to quantify and display the student's attention change trend.

[0030] The above-mentioned S31 uses Gaussian kernel density estimation to calculate the gaze density of different areas, as follows:

[0031]

[0032] Among them, (X, Y) is the image coordinate of the current probability density calculation, (X i ,Y i ) is the coordinate of the i-th fixation point, K(,) is the Gaussian kernel function, h is the smoothing parameter, and n is the number of data points.

[0033] The above S32 is as follows:

[0034] If the gaze tracking data determines that the student's gaze remains within the set radius r within the set time window Δt, it is considered as a gaze on the corresponding gaze area, and the gaze duration T is calculated. f :

[0035] T f =t end -t start (1.6)

[0036] where t end , t start are the start and end time of a fixation, respectively;

[0037] If T f If it is less than the set threshold, then the fixation is considered as a saccade and not as a valid fixation; otherwise, it is considered as a valid fixation.

[0038] Calculate the average fixation time of students on the sight area corresponding to radius r according to the effective fixation times:

[0039]

[0040] Where N is the number of effective fixations on the visual area;

[0041] Then calculate the attention stability index σ T :

[0042]

[0043] Among them, σ T The larger the value, the greater the fluctuation of students' attention.

[0044] The present invention has the following beneficial effects:

[0045] The present invention configures a high-definition, large-scene eye-tracking camera at the front of the classroom, combines an improved target tracking and high-precision pupil center extraction algorithm to analyze the gaze trajectory, can stably detect pupil position under complex lighting conditions, collect students' eye movement data in real time, and improve gaze tracking accuracy.

[0046] The present invention establishes a quantifiable teaching attention evaluation indicator, uses gaze hotspot analysis technology to quantify students' classroom attention distribution, intuitively presents students' attention distribution in the classroom, identifies high-attention areas and low-attention areas, and provides data support for teachers to adjust their teaching methods; through gaze duration statistics, it quantifies students' attention to different teaching contents, analyzes the trend of classroom attention changes, and detects teaching links that may cause students' attention to decline.

[0047] This invention establishes a quantifiable teaching attention assessment metric by collecting and analyzing student eye movement data in real time. This provides a scientific basis for teaching optimization, enabling objective and efficient classroom teaching evaluation and avoiding the subjective errors of traditional assessment methods. It can be combined with desktop intelligent camera classroom equipment, augmented reality (AR) teaching, and intelligent management systems to further enhance the level of intelligent teaching and provide technical support for the digital transformation of education.

[0048] This invention can evaluate classroom teaching effectiveness in real time and objectively, reduce subjective evaluation errors, and improve the accuracy and efficiency of teaching feedback. It is suitable for various scenarios such as traditional classrooms, online education, and experimental teaching, and has industrial application prospects for the digital transformation of education. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a schematic diagram of the application of the present invention;

[0050] Figure 2 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0052] Although the steps in the present invention are arranged with numbers, they are not intended to limit the order of the steps. Unless the order of the steps is clearly stated or the execution of a step requires other steps as a basis, the relative order of the steps can be adjusted. It is understood that the term "and / or" used herein refers to and covers any and all possible combinations of one or more of the associated listed items.

[0053] like Figure 1-2 As shown in FIG, the present invention provides a method for collecting data on attention assessment in large-scale ideological and political courses based on gaze tracking, comprising the following steps:

[0054] S1, infrared camera captures students’ eye images in real time during the course teaching process;

[0055] S2. Perform pupil area detection and pupil center fitting on the eye image to obtain gaze tracking data;

[0056] S3. Perform gaze hotspot analysis and gaze duration statistics based on gaze tracking data to quantify students’ attention distribution and changing trends and visualize them.

[0057] S4. Generate a teaching quality evaluation report and teaching optimization suggestions based on the gaze hotspot analysis and gaze duration statistics.

[0058] In the embodiment, the above method is implemented by a large-scale ideological and political course teaching attention assessment data collection system based on eye tracking, which includes an eye movement data collection module, an image processing module, a data analysis module, and a visual feedback interface;

[0059] Eye tracking module, used to collect students' eye movement data;

[0060] Image processing module, used to pre-process the collected eye movement data, including pupil area detection and ellipse fitting, to improve eye tracking accuracy;

[0061] The data analysis module is used to calculate classroom teaching quality evaluation indicators based on eye movement data, including gaze hotspot analysis and gaze duration statistics, to quantify students' attention distribution and changing trends;

[0062] The visual feedback interface is used to generate teaching quality evaluation reports based on analysis results and provide targeted teaching optimization suggestions.

[0063] In the embodiment, S1 first calibrates the high-definition, large-scale eye tracking device system to obtain system parameters; then uses an infrared light source for illumination and uses an infrared camera to capture the student's eye features in real time to obtain an eye image;

[0064] In the embodiment, S2 obtains the pupil position and gaze direction of the student through an eye tracking device, and improves the stability and accuracy of the eye movement data through steps such as image preprocessing, pupil area detection, and ellipse fitting to the pupil center. Preferably, the specific methods of image preprocessing, pupil area detection, and ellipse fitting to the pupil center are as follows:

[0065] S21. During eye tracking, the acquired eye images may be affected by factors such as illumination changes, noise interference, and eyelid occlusion. Therefore, the goal of image preprocessing is to enhance the contrast of the pupil region and improve the accuracy of subsequent pupil detection.

[0066] The eye image is converted to a grayscale image to remove color information, making subsequent edge detection and threshold segmentation more stable. Gaussian filtering is used to remove noise and enhance the edge features of the pupil area. The Gaussian filter kernel function is defined as follows:

[0067]

[0068] Where G(x,y) is the grayscale image pixel coordinate and σ controls the degree of smoothness.

[0069] S22. The pupil typically appears as a low-brightness elliptical region and can therefore be detected using edge detection and adaptive threshold segmentation methods. Canny edge detection is used to extract pupil edge feature points, and morphological operations are performed to remove noise and enhance pupil edge connectivity. The detection process includes Gaussian smoothing, gradient calculation, non-maximum suppression, and double thresholding. The formula for calculating the amplitude gradient is as follows:

[0070]

[0071] Among them, G x and G y The horizontal and vertical gradient values ​​are calculated by the Sobel operator respectively. The Otsu threshold segmentation method is used to automatically determine the optimal binarization threshold to separate the pupil area.

[0072] S23. Use an ellipse fitting algorithm to perform edge detection on the pupil area, and obtain the geometric parameters of the pupil based on the ellipse model fitting, including the ellipse center coordinates, major axis length, minor axis length and ellipse rotation angle, so as to determine the center position and shape of the pupil, and finally obtain gaze tracking data.

[0073] Ellipse fitting: The least squares ellipse fitting algorithm is used to fit the ellipse, and the center coordinates of the pupil are calculated by the parametric equation of the ellipse equation. The major and minor axes of the pupil are calculated by the radius of the ellipse, and the rotation angle is calculated by the matrix eigenvalue.

[0074] After pupil area detection, the image processing module uses an ellipse fitting algorithm to reduce the impact of illumination changes on pupil recognition and accurately locate the pupil center and shape. The ellipse is fitted using least squares to obtain the center. The fitting objective function can be expressed as:

[0075]

[0076] Where f(A,B,C,D,E,F) is the ellipse parameter set, A,B,C,D,E,F are the ellipse parameters, and the parameters of the fitted ellipse are obtained by solving the above linear equations. (x,y) are the edge point coordinates, and (x0,y0) are the pupil center coordinates.

[0077] Pupil shape determination: The flatness of the pupil is determined based on the ratio of the major axis to the minor axis of the fitted ellipse (usually the pupil is close to a perfect circle). If the ratio of the major axis to the minor axis is close to 1, the pupil shape is close to a circle, otherwise it is an ellipse.

[0078] In the embodiment, S3 is based on the collected eye movement data, and the teaching behavior analysis module uses methods such as gaze hotspot analysis and gaze duration statistics to quantify the distribution of students' classroom attention. Preferably, classroom teaching quality evaluation indicators are calculated based on eye movement data, including gaze hotspot analysis, gaze duration statistics, etc., to quantify students' attention distribution and change trends, and a classroom attention heat map is generated based on the Mean-Shift clustering algorithm to intuitively present students' attention distribution, as follows

[0079] S31. Gaze hotspot analysis is used to visualize students' attention distribution and determine which areas in the classroom are the most focused. Gaussian kernel density estimation is used to calculate the gaze density of different areas:

[0080]

[0081] Where K is the Gaussian kernel function, h is the smoothing parameter, and n is the number of data points. The Mean-Shift algorithm is used to cluster gaze concentration areas, identifying high-attention areas in the classroom. The gaze density in different areas is mapped to a color gradient, with redder colors indicating longer gaze durations and bluer colors indicating lower attention.

[0082] S32. Gaze duration refers to the time a student stays in a fixed area, reflecting the student's concentration on the content in that area. This method is used to quantify students' attention to classroom content and analyze the impact of different teaching links on attention. Set a gaze stability threshold. If the student's gaze remains within a radius r (generally set to a 1° visual angle range) within a certain time window Δt, it is considered a gaze, and the gaze duration T is calculated. f :

[0083] T f =t end -t start (1.6)

[0084] where t end , t start are the start and end time of gaze respectively. Set the gaze time threshold T f ≥100ms. Any time shorter than this is considered an eye saccade and not a valid fixation. Calculate the average fixation time in different areas (blackboard, PPT, teacher, classmates, etc.):

[0085]

[0086] Where N is the number of fixations in this area. Draw an attention trend curve on the entire class timeline to detect whether students' attention drops at certain moments and calculate the attention stability index:

[0087]

[0088] Among them, σ T The larger the value, the greater the fluctuation in students' attention and the class rhythm may need to be optimized.

[0089] In this embodiment, S4 generates a classroom quality assessment report based on the analysis results and provides teaching optimization suggestions, enabling feedback and optimization, and building a teaching attention and quality assessment model. The assessment feedback module uses data visualization technology to intuitively present the changing trends of students' attention and generate teaching optimization suggestions.

[0090] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

[0091] In addition, it should be understood that although this specification is described in terms of implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

Claims

1. A data collection method for attention assessment of large-scale ideological and political courses based on eye tracking, characterized by: The following steps are involved: S1, high-definition large-scene eye tracking photography equipment uses infrared cameras to capture students' eye images in real time during the course teaching process; S2. Perform pupil area detection and pupil center fitting on the eye image to obtain gaze tracking data; S3. Perform gaze hotspot analysis and gaze duration statistics based on gaze tracking data to quantify students’ attention distribution and changing trends and visualize them. S4. Generate a teaching quality evaluation report and teaching optimization suggestions based on the gaze hotspot analysis and gaze duration statistics.

2. The method for collecting data on attention assessment in large-scale ideological and political courses based on gaze tracking according to claim 1 is characterized in that: The S2 performs pupil area detection and pupil center fitting on the eye image to obtain eye movement data, including: S21, converting the eye image into a grayscale image, and removing noise using a Gaussian filter kernel function to obtain a denoised grayscale image G(x,y); S22. Calculate the horizontal and vertical gradient values ​​of the denoised grayscale image G(x,y) to obtain a gradient amplitude G, and automatically determine an optimal binarization threshold using the Otsu threshold segmentation method. Identify areas with a gradient amplitude greater than the threshold as pupil areas. S23. Use an ellipse fitting algorithm to perform edge detection on the pupil area, and obtain the geometric parameters of the pupil based on the ellipse model fitting, so as to determine the center position and shape of the pupil, and finally obtain the gaze tracking data.

3. The method for collecting data on attention assessment in large-scale ideological and political courses based on gaze tracking according to claim 2 is characterized in that: The Gaussian filter kernel function described in S21 is as follows: Where G(x,y) is the denoised grayscale image, x and y are the horizontal and vertical coordinates of the grayscale image pixels, and σ is the smoothness control parameter.

4. The method for collecting data on attention assessment in large-scale ideological and political courses based on gaze tracking according to claim 2 is characterized in that: The gradient calculation formula in S22 is as follows: Among them, G x and G y are the horizontal and vertical gradient values ​​of G(x,y) respectively.

5. The method for collecting data on attention assessment of large-scale ideological and political courses based on gaze tracking according to claim 2 is characterized in that: The ellipse fitting algorithm described in S23 specifically obtains the parameters of the ellipse by solving the following objective function fitting, and the corresponding ellipse center coordinates (x0, y0) are the pupil center coordinates: Where f(A,B,C,D,E,F) is the ellipse parameter set, A,B,C,D,E,F are the ellipse parameters, (x,y) are the edge point coordinates, and (x0,y0) are the pupil center coordinates.

6. The method for collecting data on attention assessment in large-scale ideological and political courses based on gaze tracking according to claim 1 is characterized in that: The S3 performs gaze hotspot analysis and gaze duration statistics based on gaze tracking data to quantify the student's attention distribution and changing trends and visualize them, including: S31. Determine the gaze area based on the gaze tracking data, and use Gaussian kernel density estimation to calculate the gaze density f(X,Y) of different areas. Cluster the gaze densities of different gaze areas to generate a classroom attention heat map to intuitively present the distribution of students' attention. S32. Determine the student's gaze area based on the gaze tracking data and identify effective gazes based on the gaze duration. Calculate the attention stability index based on the number of effective gazes to quantify and display the student's attention change trend.

7. The method for collecting data on attention assessment in large-scale ideological and political courses based on gaze tracking according to claim 6 is characterized in that: S31 uses Gaussian kernel density estimation to calculate the gaze density of different regions, as follows: Among them, (X, Y) is the image coordinate of the current probability density calculation, (X i ,Y i ) is the coordinate of the i-th fixation point, K(,) is the Gaussian kernel function, h is the smoothing parameter, and n is the number of data points.

8. The method for collecting data on attention assessment in large-scale ideological and political courses based on gaze tracking according to claim 6 is characterized in that: The S32 is specifically as follows: If the gaze tracking data determines that the student's gaze remains within the set radius r within the set time window Δt, it is considered as a gaze on the corresponding gaze area, and the gaze duration T is calculated. f : T f =t end -t start (1.6) where t end , t start are the start and end time of a fixation, respectively; If T f If it is less than the set threshold, then the fixation is considered as a saccade and not as a valid fixation; otherwise, it is considered as a valid fixation. Calculate the average fixation time of students on the sight area corresponding to radius r according to the effective fixation times: Where N is the number of effective fixations on the visual area; Then calculate the attention stability index σ T : Among them, σ T The larger the value, the greater the fluctuation of students' attention.

Images