Method and system for collecting eye movement data based on human-computer interaction

By using adaptive brightness compensation and light adaptation coefficient processing, combined with Kalman filtering technology, the problem of data discontinuity in eye-tracking systems under complex environments was solved, achieving stable eye position tracking and gaze change sequence correction, thus improving the accuracy and robustness of eye-tracking data acquisition.

CN122363527APending Publication Date: 2026-07-10XIAMEN NINE INVESTMENT TEN TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610826473.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-06-09
Publication Date
2026-07-10

AI Technical Summary

Technical Problem

Existing eye-tracking technologies suffer from discontinuous data output due to changes in lighting and user head movements in complex dynamic interactive environments, which affects the accuracy of gaze coordinate recognition.

Method used

Adaptive brightness compensation is achieved by acquiring ambient light intensity and head movement posture data. Feature extraction and trajectory reconstruction are performed by combining light adaptation coefficient and dynamic interference factor. Kalman filtering is used for eye position tracking to achieve iterative noise reduction and correction of gaze change sequence.

Benefits of technology

Under complex lighting and user head movement conditions, the accuracy and continuity of eye-tracking data acquisition were improved, the system's robustness to complex external environments was enhanced, and real-time and stable eye position tracking was achieved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122363527A_ABST
    Figure CN122363527A_ABST
Patent Text Reader

Abstract

This invention relates to the field of human-computer interaction technology, and discloses a method and system for acquiring eye-tracking data based on human-computer interaction. The method includes acquiring ambient light intensity, head movement posture data collected by a posture sensor, and a sequence of raw eye images in a human-computer interaction scenario; performing adaptive brightness compensation and feature extraction on the raw eye image sequence based on the ambient light intensity and the head movement posture data, and smoothing it to obtain smooth trajectory data; performing offset compensation and position tracking on the smooth trajectory data based on the light adaptation coefficient and the extracted dynamic interference factor to obtain stable tracking coordinates; and performing frequency domain iterative noise reduction and desensitization correction operations on the stable tracking coordinates to obtain a gaze change sequence. This method can effectively eliminate the dual physical interference of drastic fluctuations in lighting and frequent head movements in dynamic scenes, achieving high-precision and high-continuity acquisition of eye-tracking data.
Need to check novelty before this filing date? Find Prior Art