Design and Implementation of Oculogica's Eye-Tracking Device And Device Drivers for Linux

Abstract

The user's eyes constantly moves unless they makes a stop and gathers the attention onto a single point. Eye tracker information is integrated by a remotely or head-mounted Eye Tracker Device attached to a system. Whereas there are many alternative kinds of non- intrusive eye trackers, they commonly contains two common components: a infrared light supply and a camera. The light supply (infrared in our case) is directed toward the eye of the user. The camera located near user helps to track the reaction came from the corneal reactions together with visible ocular features like the pupil. This information is employed to see the rotation of the attention and ultimately the direction of gaze. Further data like blink timings and changes in pupil radius also are detected by the gaze tracker. The aggregate information is written to a file that's compatible with eye-tracking analysis module (EDF in our case).We have to design and implement a eye-tracking device to detect human brain response to the given stimuli. In eye tracking process basically what happens is human brain directs the eye where to see. So, having record of eye pattern can be useful for detecting human behavior towards the provided stimuli. In our system set-up, there are two systems provided, one for the operator (Host PC)and one for the stimuli where a patient has to look (Display PC). Both the systems operates on QNX platform and connected to each other with the LAN cable. For the Display PC, there is a camera focused on the patient's eye and a infrared light source to project light on the patient's eyes. A video stimuli will be played on display PC which has the quarter size of the screen and it will be rotating on the complete screen. Using set-up of camera and light source, the graph of the movement of retina can be plotted using light reflected from the eyes. The obtained information of the eye-pattern is then stored locally and also transferred to the cloud source using transmission control protocol (TCP). This data is then compared to some reference data which is already stored in the database and the behavior of human brain can be detected. Eye-tracking devices has many more applications like in advertising industry to detect which part of the advertise a normal viewer skips and which he follows or can be used in automobile applications to monitor if the driver fall asleep during driving. This thesis covers the implementation of Eye-Tracking Device. Primary goal of the implementation is to detect the human behavior. Earlier in this thesis, Advanced Transportation Controller part is covered. Intelligent Transportation Systems (ITS) have attracted increasing attention in recent years due to their great potential in meeting challenges like handling heavy traffic and controlling lane. We are designing a traffic controller which will reside near the traffic lights and will monitor the traffic light controlling and also several other functions like lane control. This thesis covers the implementation of ATC standard and its various peripherals.