Video Post Processing Method For On Board Vehicle Camera with Integrated Eye Tracker

This article describes how to process Eye Tracker System (ETS) data from recorded videos. The research task consisted in confirming the literature fact that the eyes of the moving person (in our case the driver in the car) inadvertently concentrate their position of the eye’s sharp vision center (ESVC) on the place on the scene where the center of the optical flow is located. ETS video records were obtained during experiments in a real vehicle test environment. As part of the post-processing of ETS videos, we determined the numerical difference between the sharp eye viewing position center and the center of the optical flow center (FOE, focus of expansion) for each recorded image. In video post processing, the vibration of the driver’s head in moving car were corrected at it was based on recorded acceleration data. The correcting of acceleration data from the ETS had significantly improved the results of the difference assessment of both centers – ESVC and FOE. A program framework was created in the MATLAB computing environment and it is ready for future use. Lack of concentration in a driver due to fatigue is a major cause of road accidents. This approach (the measurement of distance between ESVC and FOE positions) can be used in role input data generator to develop video processing and artificial intelligence based system to automatically detect driver fatigue and warn the driver, in order to prevent accidents.