The Development of Artificial Intelligence-Based Web Application to Determine the Visibility Level of the Objects on the Road

Abstract

The smallest objects that can lead to accidents on the road are called critical objects. The timely recognition of critical objects by drivers increases the safety of people driving. With a mathematical model called “visibility level” developed by Adrian in 1982, the visibility of critical objects by drivers is expressed numerically. The visibility level is defined as the ratio of the difference between the luminance of the object and the background luminance to the minimum luminance difference (luminance difference threshold) required for the object to be seen. In this model, there are many factors such as luminance of the object, background luminance, contrast polarity, observer age and the duration of the observation. Using the web application developed in this study, the visibility level of the critical object is expressed mathematically. The fundamental logic of the software consists of two stages. In the first stage, the luminance of the road and the object on the road is learned. An artificial intelligence-based algorithm has been developed to learn these luminance values. In the second stage, the observer’s features are entered into the system as the level of visibility may vary according to the features of the observer. In the second stage, the features of the observer are entered into the system, since the level of visibility will change according to the features of the observer. Once this information has been entered into the system, the software can calculate how well the object is visible to the observer.