Highway tunnel entrances and exits are accident-prone locations, especially in the daytime, primarily because of the sharp transitions in lighting in these areas. Therefore, a rational assessment of lighting transitions at tunnel portals is the key to ensuring traffic safety. This study used the EMR-8B eye-tracker system to monitor the pupil changes of eight selected drivers as they drove through 26 typical highway tunnels. The test results showed a power function relationship between drivers’ pupil areas and pupil illuminance at highway tunnel portals. A quantitative relationship between the pupil area and its critical velocity was also established, and the ratio ( k) of the pupil area's changing rate in relation to its critical velocity was used to assess the visual load and to further evaluate the lighting transitions. The results demonstrated that ( a) the relationship between pupil illuminance at the tunnel portal and the driver's pupil area conformed to Stevens’ law from experimental psychology, ( b) the relationship between the driver's pupil area and its critical velocity followed the quadratic function, ( c) the visual load at a tunnel entrance was heavier than that at the tunnel exit, and ( d) the severe transitions in pupil illuminance within 10 m of some existing highway tunnel entrances caused a great visual load, and thus are urgently in need of improvement.

The sharp variation of illumination at the entrance and exit of highway tunnels causes difficulty in the visual adapta tion of drivers and may even lead to accidents. Therefore, ascertaining the light and dark adaption time is a basic task to pro mote the safety of tunnel transportation. Took 26 typical tunnels as experimental settings and used the IViewX HED Laptop eye tracker system, the light and dark adaption of drivers in highway tunnels had been experimentally investigated. The relation be tween pupil area and its variation velocity was presented based on many experimental data and the ratio k of the variation velocity to its critical variation velocity was used to evaluate the visual load in highway tunnels and thus ascertained the visual adaption time, whereby the relation between tunnel length and visual adaptation time was thus obtained. Experimental results indicated that the dark and light adaption time of tunnels was less than 23 s and 13 s for medium and long length tunnels, respectively.

No abstract