Comprehensive Fault-Tolerant Eye-Tracking Interaction Method and System

Abstract

This invention provides a comprehensive, error-tolerant eye-tracking interaction method and system. The method includes: having a subject sit in front of a fixed eye-tracking data acquisition device in a preset scenario; collecting eye-tracking features when the subject gazes at infographics from different orientations as first gaze focus data, and collecting eye-tracking features when the subject gazes at infographics magnified to a specific magnification as second gaze focus data, while simultaneously recording the relative position information of the infographics; synchronously analyzing the number of first gaze focus data, second gaze focus data, and the relative position information of the infographics to obtain effective data; and displaying the effective data through a preset linear regression model to predict gaze focus data at different positions, angles, and trajectories when gazing at the same target hotspot, thereby increasing the probability of multiple gaze focus when gazing at the same hotspot information and improving the error tolerance of the eye-tracking human-computer interaction method.

Description

Technical Field

[0001] This document relates to the field of human-computer interaction technology, and in particular to a comprehensive fault-tolerant eye-tracking interaction method and system. Background Technology

[0002] In existing methods for human-computer sensory interaction research, eye-tracking technology can determine the pupil position and gaze direction by recording the user's gaze focus and eye movement trajectory, thereby identifying the target hotspot information that the user is paying attention to.

[0003] The human eye's field of vision is typically 124 degrees, about 25 degrees when concentrating, and between 10 and 20 degrees when accurately recognizing information. Therefore, the focal point of the gaze is always uncertain under standard visual line of sight, with 10 degrees being a sensitive area. Any information within this divergent effective visual field can become a target of human attention. Because the gaze focus data always contains a 10-degree visual cone error, the target hotspot information assessed using this method cannot be accurately matched, significantly reducing the accuracy and precision of eye-tracking human-computer interaction systems, thus resulting in a certain failure rate.

[0004] Currently, human-computer interaction eye-tracking methods all measure the user's gaze point and direction during data computation to complete gaze tracking. The more mainstream eye-tracking methods include the pupil-corneal reflection method, the corneal reflection matrix method, and the ellipse normal method. While these methods can detect eye movement features, they still have the following problems:

[0005] 1. Due to the uncertainty of gaze focus, the accuracy of eye movement feature data extracted by the device during human-computer interaction is low, and it cannot be accurately matched with the target hotspot information for evaluation;

[0006] 2. Existing eye-tracking human-computer interaction devices have low fault tolerance in their algorithms, and there are difficulties in associating the gaze focus with screen elements. Even slight deviations in the algorithm can directly affect the tracking accuracy of the interaction system.

[0007] Statistics uses methods such as searching, organizing, analyzing, and describing data to infer the essence of the object being measured and even predict its future. Therefore, statistical methods can be used to calculate the correlation between effective focal points and actual target hotspots, thereby increasing the range and probability of allowable matching errors. Summary of the Invention

[0008] To improve the tracking accuracy of eye focus in existing human-computer interaction methods under optical conditions and ensure the accuracy and effectiveness of the focused target hotspot, this invention aims to calculate the correlation between eye focus and target hotspot based on statistical laws, reduce the probability of instability caused by eye focus in the human-computer interaction system, and improve the fault tolerance of eye-tracking human-computer interaction methods.

[0009] This invention provides a comprehensive, fault-tolerant eye-tracking interaction method, comprising:

[0010] S1. In a pre-set scenario, have the subject sit in front of a fixed eye-tracking data acquisition device;

[0011] S2. Place the infographics in different positions on the subject and collect the eye movement characteristics of the subject when looking at the infographics in different positions as the first gaze focus data, and simultaneously record the relative position information of the infographics.

[0012] S3. After enlarging the text on the infographic by a specific factor, collect the eye movement characteristics of the subjects when they look at the infographic as the second gaze focus data, and simultaneously record the relative position information of the infographic;

[0013] S4. Simultaneously analyze the number of first gaze focus points, the data of second gaze focus points, and the relative position information of the information map to obtain effective data;

[0014] S5. Display the effective data through a preset linear regression model to predict the focus data of gaze at different positions, angles, and trajectories when looking at the same target hotspot.

[0015] This invention provides a comprehensive, fault-tolerant eye-tracking interaction system, comprising:

[0016] The acquisition module is used to collect eye movement feature data of subjects through an eye movement data acquisition device;

[0017] The data analysis module is used to analyze eye movement feature data and obtain effective data;

[0018] The data prediction module is used to display effective data through a preset linear regression model, thereby predicting the focus data of gaze at different positions, angles, and trajectories when looking at the same target hotspot.

[0019] This invention extracts, analyzes, and records gaze focus data such as fixation time, frequency, saccade trajectory, saccade duration, pupil size change, and blink frequency in eye movement feature detection. Multiple index analysis and comparison are used to compensate for the uncertainty of gaze focus caused by the angle of the human gaze cone. Various processing and recognition methods, including software algorithms, mechanical, electronic, and optical techniques, are combined with statistical laws to enhance the gaze focus learning algorithm for target hotspots. This reduces the failure rate of the human-computer interaction system to an acceptable level, significantly improving the accuracy and effectiveness of gaze focus tracking to target hotspot information, and providing higher tracking precision in human-computer interaction. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in one or more embodiments of this specification or in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart of a comprehensive fault-tolerant eye-tracking interaction method according to an embodiment of the present invention;

[0022] Figure 2 This is a schematic diagram of a comprehensive fault-tolerant eye-tracking interaction system according to an embodiment of the present invention. Detailed Implementation

[0023] To enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of this document.

[0024] Method Implementation Examples

[0025] This invention provides a comprehensive, fault-tolerant eye-tracking interaction method. Figure 1 This is a flowchart of a comprehensive fault-tolerant eye-tracking interaction method according to an embodiment of the present invention. Figure 1 As shown, the comprehensive fault-tolerant eye-tracking interaction method of this invention specifically includes:

[0026] S1. In a pre-set scenario, have the subject sit in front of a fixed eye-tracking data acquisition device; step S1 specifically includes:

[0027] An eye-tracking data acquisition device is fixed in the testing environment, along with an infographic for the subject to focus on a target hotspot. The subject sits in front of the eye-tracking data acquisition device to receive eye-tracking data.

[0028] S2. Place the infographics in different positions on the subject, collect the eye movement characteristics of the subject when looking at the infographics in different positions as the first gaze focus data, and simultaneously record the relative position information of the infographics; Step S2 specifically includes:

[0029] Participants were instructed to fixate on an infographic: the infographic was placed in front of the participant's left, right, and directly in front of them, with each position held for 20 seconds. This significantly altered the participant's gaze focus when focusing on the target hotspot. An eye-tracking device was used to simultaneously record the relative position of the infographic and the participant's eye movement characteristics during fixation, including fixation time, frequency, saccade trajectory, saccade duration, pupil size changes, and blink frequency.

[0030] S3. After enlarging the text on the infographic by a specific factor, collect the eye movement characteristics of the subject when looking at the infographic as the second gaze focus data, and simultaneously record the relative position information of the infographic; Step S3 specifically includes: repeating the operation of step S2, while enlarging the text on the infographic by one time and slightly changing the gaze focus angle of the subject when looking at the target hot spot.

[0031] S4. Simultaneously analyze the number of first gaze focal points, the data of the second gaze focal points, and the relative position information of the information map to obtain valid data; Step S4 specifically includes:

[0032] The relative position information of the first gaze focus number, the second gaze focus data, and the information map are analyzed and recorded synchronously. The characteristics and quantitative relationships of the number of gazes when focusing on the same target hotspot are compared longitudinally on the same time axis to find the differences in the data. An algorithm is used to perform error analysis to obtain valid data.

[0033] S5. The effective data is displayed using a preset linear regression model to predict the focus data of gaze at different positions, angles, and trajectories when focusing on the same target hotspot. Step S5 specifically includes:

[0034] Establish a regression model for gaze focus data from different locations, angles, and trajectories when the same target hotspot is being observed.

[0035] The effective data is presented as a linear regression model to predict the gaze focus data of different positions, angles, and trajectories when focusing on the same target hotspot during human-computer interaction.

[0036] By employing the embodiments of the present invention, the following beneficial effects are achieved:

[0037] In eye-tracking feature detection, data such as fixation time, frequency, saccade trajectory, saccade duration, pupil size change, and blink frequency are extracted, analyzed, and recorded. Multiple indicators are used for analysis and comparison to compensate for the uncertainty of eye focus caused by the angle of the human eye's visual cone. Various processing and recognition methods, including software algorithms, mechanical, electronic, and optical methods, are combined with statistical laws to enhance the learning algorithm of eye focus on target hotspots. This reduces the failure rate of the human-computer interaction system to an acceptable level, significantly improves the accuracy and effectiveness of eye focus tracking to target hotspot information, and provides higher tracking accuracy for human-computer interaction.

[0038] System Implementation Examples

[0039] This invention provides a comprehensive, fault-tolerant eye-tracking interactive system. Figure 2 This is a schematic diagram of a comprehensive fault-tolerant eye-tracking interaction system according to an embodiment of the present invention. Figure 2 As shown, the comprehensive fault-tolerant eye-tracking interaction system of this invention specifically includes:

[0040] The acquisition module 20 is used to acquire eye movement feature data of the subject through an eye movement data acquisition device. Specifically, the acquisition module 20 is used to: fix the eye movement data acquisition device in the test environment, and to acquire an information map of the target hot spot for the subject to gaze at. The subject sits in front of the eye movement data acquisition device to receive eye movement data acquisition.

[0041] The infographics were placed in different positions on the subjects, and the eye movement characteristics of the subjects when looking at the infographics in different positions were collected as the first gaze focus data, and the relative position information of the infographics was recorded at the same time.

[0042] After enlarging the text on the infographic by a certain factor, the eye movement characteristics of the subjects when they look at the infographic are collected as the second gaze focus data, and the relative position information of the infographic is recorded simultaneously.

[0043] Specifically, eye movement characteristics include: fixation time, number of fixations, saccade trajectory, saccade duration, pupil size changes, and blink frequency.

[0044] Data analysis module 22 is used to analyze the eye movement feature data and obtain valid data; specifically, data analysis module 22 is used for:

[0045] The relative position information of the first gaze focus number, the second gaze focus data, and the information map are analyzed and recorded synchronously. The characteristics and quantitative relationships of the number of gazes when focusing on the same target hotspot are compared longitudinally on the same time axis to find the differences in the data. An algorithm is used to perform error analysis to obtain valid data.

[0046] The data prediction module 24 is used to display effective data through a preset linear regression model, thereby predicting the focus data of gaze at different positions, angles, and trajectories when focusing on the same target hotspot. Specifically, the data prediction module 24 is used for:

[0047] Establish a regression model for gaze focus data from different locations, angles, and trajectories when the same target hotspot is being observed.

[0048] The effective data is presented as a linear regression model to predict the gaze focus data of different positions, angles and trajectories when focusing on the same target hotspot during human-computer interaction.

[0049] By employing the embodiments of the present invention, the following beneficial effects are achieved:

[0050] In eye-tracking feature detection, data such as fixation time, frequency, saccade trajectory, saccade duration, pupil size change, and blink frequency are extracted, analyzed, and recorded. Multiple indicators are used for analysis and comparison to compensate for the uncertainty of eye focus caused by the angle of the human eye's visual cone. Various processing and recognition methods, including software algorithms, mechanical, electronic, and optical methods, are combined with statistical laws to enhance the learning algorithm of eye focus on target hotspots. This reduces the failure rate of the human-computer interaction system to an acceptable level, significantly improves the accuracy and effectiveness of eye focus tracking to target hotspot information, and provides higher tracking accuracy for human-computer interaction.

[0051] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A comprehensive, fault-tolerant eye-tracking interaction method, characterized in that, include: S1. In a pre-set scenario, have the subject sit in front of a fixed eye-tracking data acquisition device; S2. Place the infographics in different positions on the subject and collect the eye movement characteristics of the subject when looking at the infographics in different positions as the first gaze focus data, and simultaneously record the relative position information of the infographics. S3. After enlarging the text on the infographic by a specific factor, collect the eye movement characteristics of the subjects when they look at the infographic as the second gaze focus data, and simultaneously record the relative position information of the infographic; S4. Simultaneously analyze the number of the first gaze focus, the data of the second gaze focus, and the relative position information of the information map to obtain valid data; S5. Display the effective data through a preset linear regression model to predict the gaze focus data at different positions, angles, and trajectories when focusing on the same target hotspot. Step S4 specifically includes: The relative position information of the first gaze focus number, the second gaze focus data, and the information map are analyzed and recorded synchronously. The characteristics and quantitative relationships of the number of gazes when focusing on the same target hotspot are compared longitudinally on the same time axis to find the differences in the data. An algorithm is used to perform error analysis to obtain valid data.

2. The method according to claim 1, characterized in that, Step S1 specifically includes: An eye-tracking data acquisition device is fixed in the testing environment, along with an infographic for the subject to focus on a target hotspot. The subject sits in front of the eye-tracking data acquisition device to receive eye-tracking data.

3. The method according to claim 1, characterized in that, The step S2, which involves placing the infographic in different locations on the subject, specifically includes: The infographics were placed in front of the subject’s left, right, and front, and the subject’s gaze lingered on each location for 20 seconds.

4. The method according to claim 1, characterized in that, The specific steps in step S2, which involve collecting eye movement characteristics of the subject when gazing at information maps in different directions, include: The study collected data on the subjects' fixation time, frequency, saccade trajectory, saccade duration, pupil size changes, and blink frequency.

5. The method according to claim 1, characterized in that, Step S5 specifically includes: Establish a regression model for gaze focus data from different locations, angles, and trajectories when the same target hotspot is being observed. The effective data is presented as a linear regression model to predict the gaze focus data of different positions, angles, and trajectories when focusing on the same target hotspot during human-computer interaction.

6. A comprehensive, fault-tolerant eye-tracking interactive system, characterized in that, include: The acquisition module is used to collect eye movement feature data of subjects through an eye movement data acquisition device; The data analysis module is used to analyze the eye movement feature data and obtain effective data; The data prediction module is used to display the effective data through a preset linear regression model, thereby predicting the gaze focus data of different positions, angles and trajectories when looking at the same target hotspot. The acquisition module is specifically used for: In a test environment, an eye-tracking data acquisition device is fixed, along with an infographic for the subject to fixate on a target hotspot. The subject sits in front of the eye-tracking data acquisition device to receive eye-tracking data. The infographics were placed in different positions on the subjects, and the eye movement characteristics of the subjects when looking at the infographics in different positions were collected as the first gaze focus data, and the relative position information of the infographics was recorded at the same time. After enlarging the text on the infographic by a certain factor, the eye movement characteristics of the subjects when they look at the infographic are collected as the second gaze focus data, and the relative position information of the infographic is recorded simultaneously. The data analysis module is specifically used for: The relative position information of the first gaze focus number, the second gaze focus data, and the information map are analyzed and recorded synchronously. The characteristics and quantitative relationships of the number of gazes when focusing on the same target hotspot are compared longitudinally on the same time axis to find the differences in the data. An algorithm is used to perform error analysis to obtain valid data.

7. The system according to claim 6, characterized in that, The eye movement characteristics collected by the acquisition module when the subject gazes at information maps in different directions specifically include: gaze duration, number of gazes, saccade trajectory, saccade duration, pupil size change, and blink frequency.

8. The system according to claim 6, characterized in that, The data prediction module is specifically used for: Establish a regression model for gaze focus data from different locations, angles, and trajectories when the same target hotspot is being observed. The effective data is presented as a linear regression model to predict the gaze focus data of different positions, angles, and trajectories when focusing on the same target hotspot during human-computer interaction.

Images