The autonomic nervous system controls the size of the pupil diameter. In particular, the pupil diameter fluctuation, which constantly change the size of the pupil under constant illuminance, is correlated with the heart rate fluctuation commonly used in autonomic function estimation. Autonomic function estimation by pupil diameter fluctuation is still under development because albino rats often used in autonomic nervous function experiments. In this study, we propose a method to detect a pupil and measure pupil diameter by image processing of albino rats under near-infrared illumination without fixing eyelids. As a result of pupillary diameter measurement accuracy evaluation, the mean error rate of the measured value and the calculated value was 0.51%, and the correlation coefficient was 0.955 (p < 0.001), suggesting the usefulness of the proposed method. In addition, we obtained the pupil diameter fluctuation under restraint load using the proposed method, and examined whether utonomic nerve function can be estimated by frequency analysis. The analysis confirmed the tendency of autonomic nervous system function observed during restraint load, suggesting that autonomic nervous function estimation by pupil diameter fluctuation may be possible.

The purpose of this study is to investigate characteristics of exit choice during evacuation in underground mall fire using virtual reality (VR) escape simulator in cooperation with thirty subjects. Each subject wore a head mounted display (HMD) experienced behavior for evacuation in a virtual underground mall developed in this study. Mobile history etc. of each subject were recorded and a questionnaire survey was conducted at the same time to grasp the reasons that the subjects moved in the virtual space, and the relationships between exit choices and the reasons were organized. The occupant density (high, medium, low) were set and the combinations of four positions and two directions of the subjects at the start of evacuation were done as the experimental conditions. A series of the experiments had shown that 1) there was a potential for using a VR escape simulator for evacuation drill in public facilities such as underground malls, 2) the start position and the direction of each subject and the occupant density had effects on their decision of moving direction and exit choice, and 3) the ratios of their decision were varied based on their conditions. And, we could consider improvement plans of the VR escape simulator.