Design Of The Bracket Type Eye Tracker And Algorithm Study

With the increase of social competition and pressure, mental health and mental disease gradually become the research hot spot.eye tracker is an important instrument of research,mainly used for the recorder in the treatment of eye movement trajectory characteristics of visual information, and is widely used in attention, visual perception, reading, schizophrenia, and other fields of research,now its price is more expensive and most for imported products.In view of this,this research intends to develop a low-cost eye movement device, in order to satisfy the demands of the use of the grassroots medical units to use.This paper studied and designed a bracket type eye tracker.The development process, mainly is the hardware construction of eye tracker,writing and implementation of the software.Hardware part of the main research work including camera system, optical system, display system and regulation and support system and building of argument.Software part of the main research work including eye diagram eigenvalue extraction algorithm research, eye calibration system and eye detection interface design and writing of the algorithm.First this article employs the pupil- corneal reflection principle of speckle tracking,designed a new pupil center and corneal reflection spot center positioning algorithm.Then in the calibration subsystem design, by adopting the method of nonlinear second order polynomial function will deduce the eye diagram coordinate system and scene one-to-one mapping relation; Then real-time eye movement data into the mapping function to find the fixation point location, identification in synchronous display images, describe the eye movement trajectory.The last eye detection interface is designed, mainly include eye diagram displaying frame, scene graph displaying box, calibration, eye movement data displaying part, testing part and the medical records management, etc.The interface realized from the beginning of image acquisition to the end of eye movement trajectory of the display and eye movement testing report generate and print,also include the detailed information of the subjects and query function.Eye tracker designed in this paper completed the eye tracking experiment, the real-time display the eye image positioning effect and fixation point position are shown in the image of the scene, describe the eye movement trajectory.Although eye tracker designed in this paper completed the eye test, but the follow-up still need for hardware part and eye location algorithm will be improved.Hardware needs to adopt auto-focus infrared cameras in order to increase test efficiency, software aspects need to further improve the accuracy of the eye location algorithm and the perfection of eye detection interface.