Visual information of vehicle velocity acquired by pedestrians involved in road crossing accidents

• Published in: Accident analysis and prevention Vol. 151; p. 105912
• Main Authors: Yokoya, Yasushi, Soma, Hitoshi
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.03.2021
• Subjects: Accident statistics; Geometrical optics; Motion perception; Pedestrian accident; Retinal image; Geometrical optics; Accident statistics; Retinal image; Pedestrian accident; Motion perception
• ISSN: 0001-4575; 1879-2057
• DOI: 10.1016/j.aap.2020.105912

•Visual model based on geometric optics for the velocity of an approaching vehicle and the retinal image of a pedestrian is constructed.•Visual model gives visual information pertaining to the error in judgment made by a pedestrian crossing the road regarding the vehicle velocity.•Perceivable distance of vehicle velocity is determined by the visual information derived from either pedestrian eye height or vehicle height.•Perceivable distance of vehicle velocity is consistent with the timing a vehicle collides with a pedestrian extracted from accident statistics. This paper describes a geometric optical relationship between the perceived visual information of approaching vehicles by pedestrians who intend to cross the road and the factors underlying pedestrian accidents in Japan. We create a model based on this visual information, wherein the retinal image corresponding to vehicle velocity perception is broken down into tangential components and normal components in a two-dimensional polar coordinate system that employs the nodal point of the eyeball as the origin. Our visual model uses the relationship between the tangential and normal components of the velocity to calculate the distance at which the velocity of the vehicle can be perceived by pedestrians. The maximum distance at which vehicle velocity can be perceived by pedestrians derived from the visual model is consistent with the timing at which a vehicle collides with a pedestrian most frequently in the pedestrian accidents gleaned from accident statistics. The result of the simulation of the visual model showed that the eye height of the pedestrian, the total height of the vehicle and the sensory threshold of motion determine the components by which the maximum perception distance of the vehicle velocity is given. These findings contribute to the enhancement of safety measures in traffic accidents from the pedestrian’s perspective.