Research On Data Fusion Of Pilots’ Gaze Information In Human Computer Interaction

As the application of object detection and line of sight tracking in computer vision is more and more extensive,the technology needs to develop and improve the accuracy of detection and tracking in work and life.Its application has gradually penetrated into all aspects of people’s life,and its importance has become increasingly prominent.It has begun to occupy a great application prospect in many fields,such as human-computer interaction,fatigue detection,aviation Air traffic,etc.At the same time,with the progress of science and technology,as well as the growing demand of the people,human-computer interaction technology has been widely used.Its application is not limited to the technology involved in the manual control operation,but more intelligent,humanized and interactive human-computer interaction technology applications,such as voice,pupil,face,expression recognition and tracking positioning It can meet people’s needs,bring people more humanistic experience,and accelerate the development of society.Gaze information extraction and gaze tracking data fusion technology is the main technical tool to measure the target feature of gaze area,which is divided into three main parts:target feature extraction,gaze tracking and data fusion.After the information collected by each channel is filtered through data processing,the effective information is fused by using data fusion technology to assist the determination,verification and verification of the pilot’s gaze information,and prompt the abnormal information in time.Aiming at the problem that the accuracy of gaze information data extraction is not high in the flight process of pilots,this paper puts forward the research on the data fusion of gaze impact point and target information,which mainly includes three parts:gaze impact point estimation,data fusion and target information acquisition.Through human-computer interaction,the information is fed back to obtain accurate target information.Firstly,the non wearable eye tracker is used to track the line of sight,estimate the fixation point,and process the fixation data to obtain the effective fixation point coordinate data.Secondly,the effective fixation point is recorded as attention information,which is fused with the reaction time information.A large amount of data is used as input data,which is matched with the position coordinate of the instrument panel to get the dial where the fixation point is located.Then,data fusion is performed between the instrument panel where the fixation point is located and the flight parameters returned by multiple channels to obtain the information displayed by the instrument panel.Finally,the target information obtained is used for data support of identifying the pilot’s intention in the future,predicting the decision that the pilot will make,and prompting the decision,so as to improve the decision-making efficiency and accuracy.