Eye tracking method for human-computer interaction of mobile device

Abstract

The invention discloses an eye tracking method for human-computer interaction of a mobile device. The system is moderately comfortable, and has high precision and robustness, and high feasibility and effectiveness. The system comprises an eye image processing layer, an eye feature detection layer, an eye data calculation layer and an eye interactive application layer. The eye image processing layer comprises image acquisition and image preprocessing, namely, filtering or binarization. The eye feature detection layer comprises pupil detection, pupil location, pupil edge fitting, Pu Chin spot detection and template matching. The eye data calculation layer comprises point of fixation coordinates, pupil-conea reflection vector, point of fixation calibration, other data and pupil diameter. The eye interactive application layer comprises region of interest selection or switching, and specific interactive application.

Description

technical field [0001] The invention belongs to the field of automatic irrigation of facility agriculture, and in particular relates to an eye tracking method for human-computer interaction of mobile devices. Background technique [0002] Eyetracking, also known as gaze tracking, is used to measure the user's gaze point or gaze direction. On the one hand, eye tracking can analyze the user's visual behavior and cognitive activities, such as psychological [0003] On the other hand, it can also be used as a new input technology to replace the mouse and keyboard, such as eye-controlled typing system and eye-movement-driven adaptive user interface system. However, these applications are based on the fixed eye tracking system in the laboratory, which cannot support the analysis of eye tracking data and related human-computer interaction applications in a real environment (non-laboratory mobile environment). For example, in a conference scene, users are equipped with a mobile ey...

AI Technical Summary

Problems solved by technology

However, there are still deficiencies in this method: 1) In order to obtain the differential images of bright pupils and dark pupils, the control process of the system for the infrared light source is complicated, and the structure of the infrared light source is complicated and the volume is large, so it can only be designed as a fixed desktop Eye movement system; 2) The detection algorithm for Pu’er chin spots is relatively simple, without considering that in practical applications, Pu’er chin spots often produce deformation, loss, etc., resulting in low robustness of the algorithm

It can be seen that the existing research still has problems such as low data sampling accuracy, large equipment volume and quality, and it is difficult to support mobile interactive applications.

[0005] Traditional eye tracking devices are complex in structure, bulky and heavy, and usually can only be used in a fixed way on a desktop, and cannot support mobile interaction in a pervasive computing environment

Images